Generation of Synthetic Clinical Text: A Systematic Review

Ahmed Abdelmonem Hemedan; Basel Alshaikhdeeb; Irina Balaur; Soumyabrata Ghosh; Venkata Satagopam

REVIEW 3 major objections 6 minor 1 cited by

Generation of Synthetic Clinical Text: A Systematic Review

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Synthetic clinical text generation is dominated by transformer models, especially GPTs, and is evaluated most often by downstream utility, according to a systematic review of 94 studies.

desk verdict A useful first map of synthetic clinical free-text generation, but the inclusion criteria are too loose and the counts don't add up; qualitative findings credible, quantitative distributions not. read the letter →

arxiv 2507.18451 v1 pith:RQCCTZUX submitted 2025-07-24 cs.CL cs.AI

Basel Alshaikhdeeb , Ahmed Abdelmonem Hemedan , Soumyabrata Ghosh , Irina Balaur , Venkata Satagopam This is my paper · ORCID

classification cs.CLcs.AI

keywords syntheticclinicaltextmedicalfree-textgenerationtransformerarchitecturesGPTdataaugmentationprivacy-preservingsystematicreviewnaturallanguageprocessing

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper surveys the growing literature on generating synthetic medical free-text and organizes it into a quantitative taxonomy. It finds that the field took off after 2018, that transformer architectures—particularly GPT-style models—are the dominant generation technique, and that evaluation focuses most heavily on utility in downstream natural-language-processing tasks. The review also identifies six main generation purposes, led by data augmentation and assistive writing, and four evaluation aspects, with privacy the least settled. The authors argue that synthetic text works well as a complement to real records, improving accuracy and easing data sparsity, but that privacy risks remain a major open problem.

What carries the argument

The organizing mechanism is the systematic-review protocol with three research questions, which yields a classification scheme across purpose, technique, and evaluation dimensions. The load-bearing components are the taxonomy itself and the quantitative distributions computed from the 94-article corpus, including the six purpose categories, the four technique families with a transformer/GPT emphasis, and the four evaluation aspects with utility most common.

What would settle it

Re-run the database search with broader term placement (searching abstract and full text, not only titles), include non-English and non-open-access publications, and have two independent screeners with inter-rater reliability; if the resulting set of eligible studies yields a substantially different distribution of generation techniques or evaluation purposes (for example, transformers no longer dominant), the paper's central quantitative claims would be falsified.

Watch

Extended reading notes

Core claim

Based on 94 articles selected from 1,398 retrieved from seven databases, the paper establishes a three-part taxonomy of synthetic medical free-text generation: purposes (privacy-preserving, augmentation, usefulness, assistive writing, annotation, corpus building), techniques (manual, text processing, knowledge-based, and neural network models), and evaluation methods (similarity, privacy, structure, and utility). Quantitatively, transformer architectures, especially the GPT family, account for the largest share of generation techniques, and utility-based evaluation—measured through downstream NLP tasks such as named entity recognition, relation extraction, and report generation—is the most frequent assessment paradigm. The review further concludes that synthetic text can moderately substitute for real clinical text and strongly complements it for augmentation, while privacy evaluation is underdeveloped and human assessment for sensitive information is rarely performed.

Load-bearing premise

The completeness and representativeness of the 94-article sample: the search restricted some queries to title fields, excluded non-English and non-open-access articles, and screening was done without inter-rater reliability, so the reported distributions could be biased if the sample is not representative.

Editorial extensions

If this is right

Research effort in synthetic clinical text is likely to concentrate on GPT-style models and their prompt-based variants, leaving non-neural and older neural approaches underexplored.
Evaluation practice will need to shift toward privacy-specific metrics and human assessment, since the review finds that similarity and utility are heavily used but privacy remains the weakest evaluation aspect.
Synthetic text will be adopted primarily as augmented, complementary training data for downstream NLP tasks rather than as a full replacement for real clinical documents.
The lack of unified evaluation metrics across studies will impede direct comparison of generation techniques, motivating future empirical benchmarking.
The taxonomy provides a checklist for new work, making it easier to position a new method by purpose, technique, and evaluation paradigm.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The dominance of GPTs may partly reflect the search window (2015 to August 2024) coinciding with the rapid rise of large language models, so a re-run of the review after a few more years could shift the technique distribution toward newer open-weight models such as Llama and Mistral, which already appear in the dataset.
A natural testable extension would be to re-run the search without title-field restrictions and including non-English and non-open-access articles, then compare the resulting distributions to see whether the reported transformer dominance and utility-evaluation frequency hold.
The paper's finding that utility is the most frequent evaluation aspect while privacy is the most concerning suggests that privacy-preserving generation and its verification may be the bottleneck for clinical deployment; a focused study on membership-inference and re-identification metrics would sharpen this picture.
The taxonomy could be operationalized as a benchmark protocol where each new synthetic-text method is required to report at least one metric from each of the four evaluation aspects, which the authors' own limitation section implies is currently missing.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

First, the useful part. This is the first systematic review I know of that specifically covers synthetic clinical free-text generation, and it gives a workable taxonomy: purposes (augmentation, assistive writing, corpus building, privacy, annotation, usefulness), generation techniques (manual, text processing, knowledge-based, neural nets), and evaluation aspects (similarity, privacy, structure, utility). The qualitative conclusions—GPTs dominate, utility is the most common evaluation mode, augmentation is the strongest use case—are consistent with the literature and feel right. The appendix tables are a decent starting point for anyone entering the area.

The soft spots are real, though. The headline numbers do not reconcile: the paper says 94 included articles, but Appendix D lists 93, and several Figure 6 counts don't match the appendix entries. That alone would make me cautious about any quantitative claim.

More substantively, the inclusion criteria are too broad. The review's stated scope is text 'inferred from EMR/EHR datatypes,' but the appendix includes multiple studies that generate reports from chest X-rays or other images (e.g., [69,70,72,77,79,81,82,86,87,88]) and studies that summarize doctor-patient dialogues (e.g., [64,74,83,107]). Those are standard clinical NLG tasks, not synthetic data generation in the privacy/augmentation sense. Including them inflates the transformer count, the 'assistive writing' purpose, and the 'utility via report generation' evaluation. The authors acknowledge in Section 4.6 that they tried to stick to studies that made changes to the text, which is inconsistent with pulling in image-to-text systems. This is a construct validity issue, not just a missing comma: the quantitative answers to RQ1 and RQ3 are biased by the coding scheme.

That said, the central qualitative patterns may survive a tighter re-coding. GPTs would still be widespread, and utility-based evaluation is probably still common. But the paper cannot be taken as a trustworthy quantitative map until the counts are fixed and the inclusion rules are tightened. The search itself is broad and well documented, and the lack of inter-rater reliability on screening is a minor but real omission.

Who is this for? Researchers starting in synthetic clinical text who want a quick survey of purposes and methods, and people working on evaluation benchmarks. It deserves peer review—it's a real contribution—but it needs major revision before the numbers can be relied on. I'd recommend the editor send it out, and ask the authors to re-code with a stricter definition of synthetic text, reconcile all counts, and report screening reliability.

Referee Report

3 major / 6 minor

Summary. This manuscript presents a systematic review of synthetic clinical free-text generation. The authors searched seven databases, screened 1,398 records, and included 94 articles (per the abstract and Section 2.5; Appendix D lists 93). They pose three research questions concerning generation purpose, technique, and evaluation, and answer them with taxonomies and quantitative summaries (Figures 2–6, Appendices A–D). The headline findings are that text augmentation and assistive writing are the most frequent purposes, transformer-based models (especially GPT variants) dominate techniques, and utility-based evaluation is the most common approach, with privacy remaining a central concern.

Significance. If the quantitative claims were reliable, this would be a valuable survey of an emerging area and the first review dedicated specifically to unstructured synthetic clinical text. The paper's strengths include a detailed per-article coding table (Appendix D), explicit taxonomies for purposes, techniques, and evaluation methods, and a clear framing of three research questions. The qualitative observation that GPT-style models and utility evaluation are prominent is likely robust. However, the internal inconsistencies in article counts and the overly broad inclusion criteria undermine the quantitative distributions; the paper therefore needs a substantial but well-defined revision rather than a rejection.

major comments (3)

[2.2 / Appendix D] The inclusion criterion in Section 2.2 requires that 'the generated synthetic text should be inferred from EMR/EHR datatypes,' but Appendix D includes numerous studies that generate text from medical images or from human dialogue, e.g., [69,70,77,79,81,82,86,87,88,101,53] (image-to-text report generation) and [64,74,83,107] (dialogue summarization). These are standard clinical NLG tasks, not synthetic data generation in the privacy/augmentation sense that motivates the review. Section 4.6 states that the authors 'tried to adhere to the studies that attempted to make changes on text during the automation,' which is inconsistent with including image-to-text generation. Re-coding the corpus under a stricter definition could materially change the counts for RQ1 (purpose) and RQ3 (evaluation), and in particular the headline that GPT-style transformers are dominant and that utility is the most frequent evaluation mode.
[2.5 / Figure 6 / Appendix D] The paper reports 94 included articles in the abstract and Section 2.5, but Appendix D lists only 93 unique entries, and the year counts in Figure 6(a) sum to 93. At the same time, the purpose counts in Figure 6(c) sum to 94, and the purpose counts do not match the multi-label purpose assignments in Appendix D (e.g., [55] is coded as both Privacy-preserving and Augmentation, yet Figure 6(c) appears to assign each article to a single purpose). These discrepancies make the quantitative answers to RQ1–RQ3 unreproducible. The counts should be reconciled and the coding rules clarified.
[2.4 / 4.6] The review does not report dual screening or inter-rater reliability for title/abstract and full-text screening, and it does not describe how coding disagreements would have been resolved. Because the inclusion decision is central to every quantitative claim, the absence of such information, combined with the acknowledged selection bias of the search (Section 4.6), leaves the sample's completeness and representativeness uncertain. The authors should either provide evidence of screening reliability or temper the quantitative claims to clearly reflect this uncertainty.

minor comments (6)

[Appendix D, row [29]] The language 'Norwagian' is a typo for 'Norwegian'.
[Appendix D, row [80]] In the Evaluation Paradigm column, 'Similarly' should be 'Similarity'.
[Figure 4] The label 'Knolwedge-base' should be 'Knowledge-based'.
[Appendix C.1.1, Eq. (5)] Equation (5) writes BLEU = BP ∗ N; the factor N should be the geometric mean precision defined in Equation (3), not the n-gram order.
[Section 4.6] The phrase 'the polysemous of the term synthetic' should read 'the polysemy of the term synthetic'.
[Figure 6(c)] The paper should state explicitly whether the purpose categories in Figure 6(c) are mutually exclusive; several Appendix D entries list more than one purpose, so the counting rule needs to be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the review's claims are descriptive summaries of an external literature, not derived from fitted parameters or self-citation chains.

full rationale

This is a systematic review, not a predictive or first-principles derivation. The central claims—purpose distribution, technique dominance (especially GPT/transformers), and evaluation-method frequencies—are explicitly computed from 94 external publications enumerated in Appendix D and summarized in Section 3 and Figure 6. No equation is fitted to data and then renamed as a prediction; no result is defined into existence by an assumption. The paper's taxonomy (purposes, techniques, evaluation aspects) is an interpretive coding scheme, but the coding is applied to independent literature rather than justified by the authors' own prior work, and no uniqueness theorem or self-citation is invoked to forbid alternative classifications. The authors disclose potential selection bias and search limitations in Section 4.6, but a sampling or construct-validity concern is a correctness risk, not circularity. Therefore, no circular step can be exhibited, and the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The review has no free parameters and does not introduce any new entities. It relies on three domain assumptions about search completeness, screening reliability, and taxonomy validity. The authors themselves flag the first of these in Section 4.6, and the internal count inconsistencies weaken the credibility of the second and third.

assumptions (3)

domain assumption The deployed search strings and database set retrieve all or most relevant publications on synthetic clinical free-text generation.
Used throughout Section 2.1; the first query restricts roots to the title field and the authors explicitly acknowledge possible selection bias in Section 4.6.
domain assumption Single-reviewer title/abstract/full-text screening without inter-rater agreement yields reliable inclusion decisions.
The screening process in Section 2.5 and Figure 1 does not report dual coding or inter-rater reliability, which is a common expectation for systematic reviews.
domain assumption The proposed taxonomy (six purposes, four technique classes, four evaluation aspects) is a valid and consistently applicable coding scheme for the literature.
Applied in Section 3 and Appendices A-C; the overlapping reference lists in Figure 2 and the mismatched counts in Figure 6(c) suggest the coding scheme is not perfectly reliable.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Generation of Synthetic Clinical Text: A Systematic Review." pith.science (2026). https://pith.science/paper/RQCCTZUX

@misc{pith2026250718451,
  author       = {Pith},
  title        = {Pith review of: Generation of Synthetic Clinical Text: A Systematic Review},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RQCCTZUX}},
  note         = {Machine review of arXiv:2507.18451}
}

read the original abstract

Generating clinical synthetic text represents an effective solution for common clinical NLP issues like sparsity and privacy. This paper aims to conduct a systematic review on generating synthetic medical free-text by formulating quantitative analysis to three research questions concerning (i) the purpose of generation, (ii) the techniques, and (iii) the evaluation methods. We searched PubMed, ScienceDirect, Web of Science, Scopus, IEEE, Google Scholar, and arXiv databases for publications associated with generating synthetic medical unstructured free-text. We have identified 94 relevant articles out of 1,398 collected ones. A great deal of attention has been given to the generation of synthetic medical text from 2018 onwards, where the main purpose of such a generation is towards text augmentation, assistive writing, corpus building, privacy-preserving, annotation, and usefulness. Transformer architectures were the main predominant technique used to generate the text, especially the GPTs. On the other hand, there were four main aspects of evaluation, including similarity, privacy, structure, and utility, where utility was the most frequent method used to assess the generated synthetic medical text. Although the generated synthetic medical text demonstrated a moderate possibility to act as real medical documents in different downstream NLP tasks, it has proven to be a great asset as augmented, complementary to the real documents, towards improving the accuracy and overcoming sparsity/undersampling issues. Yet, privacy is still a major issue behind generating synthetic medical text, where more human assessments are needed to check for the existence of any sensitive information. Despite that, advances in generating synthetic medical text will considerably accelerate the adoption of workflows and pipeline development, discarding the time-consuming legalities of data transfer.

Figures

Figures reproduced from arXiv: 2507.18451 by the authors.

**Figure 1.** Selection process flow diagram pus4 (RRID: SCR 022613), IEEE5 (RRID: SCR 008314), Google Scholar6 (RRID: SCR 008878), and arXiv7 (RRID: SCR 005488). These search engines would provide a wide coverage of general scientific publications. On the other hand, we have set three search limits, including the search term, date, and type of publication. First, for the search terms, since our review focuses on synthetic text … view at source ↗

**Figure 2.** Generation Purposes Data Source Private EHR/EMR [16, 18, 20, 26, 28–30, 32, 34, 39, 42, 44, 45, 53–55, 60, 64, 90, 94] Human Collection & Curation [27, 41, 50, 65, 66] Publicly Available [19, 21, 25, 31, 33, 41, 43, 46, 48, 52, 56–58, 62, 67– 70, 72–83, 85–89, 91–93, 95–102, 104–107, 110] MIMIC-III [41, 46, 48, 52, 56– 58, 61, 62, 68, 75, 80, 90, 95, 97, 100, 107] MIMIC-CXR [25, 72, 79, 81, 86–88] IU X-RAY [69, 70, … view at source ↗

**Figure 3.** Datasets 7 [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Generation Techniques 8 [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Evaluation Methods 9 [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

**Figure 6.** Figure 6: General Statistics 12 [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

**Figure 7.** Figure 7: Neural Network Architectures 32 [PITH_FULL_IMAGE:figures/full_fig_p032_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Clinical Communication Processing with Models Trained on LLM-Generated Synthetic Data: A Structured Survey and Novel Application Case Studies
cs.CL 2026-08 conditional novelty 6.0 of 10

Synthetic clinical communication generated by LLMs can train clinical NLP models in thirteen case studies, but only one is tested on real patient text, leaving transfer to authentic communication unproven.

Reference graph

Works this paper leans on

256 extracted references · 78 canonical work pages · cited by 1 Pith paper

[55]

Scott H. Lee. Natural language generation for electronic health records. npj Digital Med , 1(1), nov 2018

2018
[1]

Should free-text data in electronic medical records be shared for research? a citizens’ jury study in the UK

Elizabeth Ford, Malcolm Oswald, Lamiece Hassan, Kyle Bozentko, Goran Nenadic, and Jackie Cassell. Should free-text data in electronic medical records be shared for research? a citizens’ jury study in the UK. J Med Ethics , 46(6):367–377, may 2020

2020
[2]

Synonym-based text generation in restructuring imbalanced dataset for deep learning models

Febi Siti Sutria Ningsih, Purnomo Husnul Khotimah, Andria Arisal, Andri Fachrur Rozie, Devi Munandar, Dianadewi Riswantini, Ekasari Nugraheni, Wiwin Suwarningsih, and Dian Kurniasari. Synonym-based text generation in restructuring imbalanced dataset for deep learning models. In 2022 5th International Conference on Networking, Information Systems and Secur...

2022
[3]

Generating natural language adversarial examples on a large scale with generative models, 2020

Yankun Ren, Jianbin Lin, Siliang Tang, Jun Zhou, Shuang Yang, Yuan Qi, and Xiang Ren. Generating natural language adversarial examples on a large scale with generative models, 2020

2020
[4]

Synthetic data generation: State of the art in health care domain

Hajra Murtaza, Musharif Ahmed, Naurin Farooq Khan, Ghulam Murtaza, Saad Zafar, and Ambreen Bano. Synthetic data generation: State of the art in health care domain. Computer Science Review, 48:100546, 2023

2023
[5]

Synthetic data generation for tabular health records: A systematic review

Mikel Hernandez, Gorka Epelde, Ane Alberdi, Rodrigo Cilla, and Debbie Rankin. Synthetic data generation for tabular health records: A systematic review. Neurocomputing, 493:28–45, 2022

2022
[6]

Deep generative models for synthetic data: A survey

Peter Eigenschink, Thomas Reutterer, Stefan Vamosi, Ralf Vamosi, Chang Sun, and Klaudius Kalcher. Deep generative models for synthetic data: A survey. IEEE Access, 11:47304–47320, 2023

2023
[7]

A systematic literature review: deep learn- ing techniques for synthetic medical image generation and their applications in radiotherapy

Moiz Khan Sherwani and Shyam Gopalakrishnan. A systematic literature review: deep learn- ing techniques for synthetic medical image generation and their applications in radiotherapy. Frontiers in Radiology, 4:1385742, 2024

2024

Show all 256 references

[8]

Ghosheh, Jin Li, and Tingting Zhu

Ghadeer O. Ghosheh, Jin Li, and Tingting Zhu. A survey of generative adversarial networks for synthesizing structured electronic health records. ACM Comput. Surv. , 56(6), January 2024

2024
[9]

Evalua- tion of synthetic electronic health records: A systematic review and experimental assessment

Emmanuella Budu, Kobra Etminani, Amira Soliman, and Thorsteinn R¨ ognvaldsson. Evalua- tion of synthetic electronic health records: A systematic review and experimental assessment. Neurocomputing, 603:128253, 2024

2024
[10]

Pezoulas, Dimitrios I

Vasileios C. Pezoulas, Dimitrios I. Zaridis, Eugenia Mylona, Christos Androutsos, Kosmas Apostolidis, Nikolaos S. Tachos, and Dimitrios I. Fotiadis. Synthetic data generation methods in healthcare: A review on open-source tools and methods. Computational and Structural Biotech...

2024
[11]

Primer on generative artificial intelligence and large language models in medical imaging

Kiduk Kim, Gil-Sun Hong, and Namkug Kim. Primer on generative artificial intelligence and large language models in medical imaging. Journal of the Korean Society of Radiology , 85(5):848–860, 2024

2024
[12]

Generative ai for synthetic 14 data across multiple medical modalities: A systematic review of recent developments and challenges

Mahmoud Ibrahim, Yasmina Al Khalil, Sina Amirrajab, Chang Sun, Marcel Breeuwer, Josien Pluim, Bart Elen, G¨ okhan Ertaylan, and Michel Dumontier. Generative ai for synthetic 14 data across multiple medical modalities: A systematic review of recent developments and challenges. ...

2025
[13]

Five steps to conducting a systematic review, 2003

Khalid S Khan, Regina Kunz, Jos Kleijnen, and Gerd Antes. Five steps to conducting a systematic review, 2003

2003
[14]

Systematic reviews and meta-analyses, February 2011

Lindsay S Uman. Systematic reviews and meta-analyses, February 2011

2011
[15]

Gomez, Lukasz Kaiser, and Illia Polosukhin

Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need, 2017

2017
[16]

Automatic generation of electronic medical record based on gpt2 model

Junkun Peng, Pin Ni, Jiayi Zhu, Zhenjin Dai, Yuming Li, Gangmin Li, and Xuming Bai. Automatic generation of electronic medical record based on gpt2 model. In 2019 IEEE International Conference on Big Data (Big Data) , pages 6180–6182, 2019

2019
[17]

Cmed-gpt: Prompt tuning for entity-aware chinese medical dialogue generation, 2023

Zhijie Qu, Juan Li, Zerui Ma, and Jianqiang Li. Cmed-gpt: Prompt tuning for entity-aware chinese medical dialogue generation, 2023

2023
[18]

Generation of synthetic elec- tronic medical record text

Jiaqi Guan, Runzhe Li, Sheng Yu, and Xuegong Zhang. Generation of synthetic elec- tronic medical record text. In 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, dec 2018

2018
[19]

Medconqa: Medical conversational question answering system based on knowledge graphs

Fei Xia, Bin Li, Yixuan Weng, Shizhu He, Kang Liu, Bin Sun, Shutao Li, and Jun Zhao. Medconqa: Medical conversational question answering system based on knowledge graphs. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstr...

2022
[20]

A method for generating synthetic electronic medical record text

Jiaqi Guan, Runzhe Li, Sheng Yu, and Xuegong Zhang. A method for generating synthetic electronic medical record text. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 18(1):173–182, January 2021

2021
[21]

Research on text generation of medical intelligent question and answer based on bi-lstm and neural network technology

Keyi Huang, Fenda Ji, Wei Lu, and Yue Xiao. Research on text generation of medical intelligent question and answer based on bi-lstm and neural network technology. In 2022 IEEE/ACIS 22nd International Conference on Computer and Information Science (ICIS) , pages 54–59, 2022

2022
[22]

Schapranow

Florian Borchert, Christina Lohr, Luise Modersohn, Thomas Langer, Markus Follmann, Jan Philipp Sachs, Udo Hahn, and Matthieu-P. Schapranow. Ggponc: A corpus of german medical text with rich metadata based on clinical practice guidelines, 2020

2020
[23]

GRASCCO — the first publicly shareable, multiply-alienated german clinical text corpus

Luise Modersohn, Stefan Schulz, Christina Lohr, and Udo Hahn. GRASCCO — the first publicly shareable, multiply-alienated german clinical text corpus. In Studies in Health Tech- nology and Informatics . IOS Press, aug 2022

2022
[24]

Sharing copies of synthetic clinical corpora without physical distribution — a case study to get around IPRs and privacy constraints fea- turing the German JSYNCC corpus

Christina Lohr, Sven Buechel, and Udo Hahn. Sharing copies of synthetic clinical corpora without physical distribution — a case study to get around IPRs and privacy constraints fea- turing the German JSYNCC corpus. In Proceedings of the Eleventh International Conference on Lan...

2018
[25]

Reinforcement 15 learning with imbalanced dataset for data-to-text medical report generation

Toru Nishino, Ryota Ozaki, Yohei Momoki, Tomoki Taniguchi, Ryuji Kano, Norihisa Nakano, Yuki Tagawa, Motoki Taniguchi, Tomoko Ohkuma, and Keigo Nakamura. Reinforcement 15 learning with imbalanced dataset for data-to-text medical report generation. In Trevor Cohn, Yulan He, and...

2020
[26]

A practical and universal framework for generating publicly available medical notes of authentic quality via the power of crowds

Rina Kagawa, Yukino Baba, and Hideo Tsurushima. A practical and universal framework for generating publicly available medical notes of authentic quality via the power of crowds. In 2021 IEEE International Conference on Big Data (Big Data) , pages 3534–3543, 2021

2021
[27]

Cognitive assessment of japanese older adults with text data augmentation

Toshiharu Igarashi and Misato Nihei. Cognitive assessment of japanese older adults with text data augmentation. Healthcare, 10(10):2051, oct 2022

2022
[28]

Instruction-guided deidentification with synthetic test cases for norwegian clinical text

Jørgen Aarmo Lund, Karl Øyvind Mikalsen, Joel Burman, Ashenafi Zebene Woldaregay, and Robert Jenssen. Instruction-guided deidentification with synthetic test cases for norwegian clinical text. In Tetiana Lutchyn, Ad ´ ın Ram ´ ırez Rivera, and Benjamin Ricaud, editors, Proceed...

2024
[29]

Iterative development of family history annotation guidelines using a synthetic corpus of clinical text

Taraka Rama, P ˚ al Brekke, Øystein Nytrø, and Lilja Øvrelid. Iterative development of family history annotation guidelines using a synthetic corpus of clinical text. In Proceedings of the Ninth International Workshop on Health Text Mining and Information Analysis . Associatio...

2018
[30]

Brekke, Taraka Rama, Ildik´ o Pil´ an, Øystein Nytrø, and Lilja Øvrelid

P ˚ al H. Brekke, Taraka Rama, Ildik´ o Pil´ an, Øystein Nytrø, and Lilja Øvrelid. Synthetic data for annotation and extraction of family history information from clinical text. J Biomed Semant, 12(1), jul 2021

2021
[31]

Can synthetic text help clinical named entity recognition? a study of electronic health records in French

Nicolas Hiebel, Olivier Ferret, Karen Fort, and Aur´ elie N´ ev´ eol. Can synthetic text help clinical named entity recognition? a study of electronic health records in French. In Andreas Vlachos and Isabelle Augenstein, editors, Proceedings of the 17th Conference of the Europ...
[32]

Generating synthetic training data for supervised de-identification of electronic health records

Claudia Alessandra Libbi, Jan Trienes, Dolf Trieschnigg, and Christin Seifert. Generating synthetic training data for supervised de-identification of electronic health records. Future Internet, 13(5):136, may 2021

2021
[33]

Synthetic Arabic medical dialogues using advanced multi-agent LLM techniques

Mariam ALMutairi, Lulwah AlKulaib, Melike Aktas, Sara Alsalamah, and Chang-Tien Lu. Synthetic Arabic medical dialogues using advanced multi-agent LLM techniques. In Nizar Habash, Houda Bouamor, Ramy Eskander, Nadi Tomeh, Ibrahim Abu Farha, Ahmed Ab- delali, Samia Touileb, Injy...

2024
[34]

AI-driven approach for automatic synthetic patient status corpus generation

Boris Velichkov, Kristina Ivanova, Valeri Hristov, Ivan Borisov, Alexander Peychev, Ivan Koychev, and Svetla Boytcheva. AI-driven approach for automatic synthetic patient status corpus generation. In 2020 4th International Conference on Artificial Intelligence and Virtual Real...

2020
[35]

Johnson, Tom J

Alistair E.W. Johnson, Tom J. Pollard, Lu Shen, Li wei H. Lehman, Mengling Feng, Moham- mad Ghassemi, Benjamin Moody, Peter Szolovits, Leo Anthony Celi, and Roger G. Mark. MIMIC-III, a freely accessible critical care database. Scientific Data , 3(1), May 2016

2016
[36]

Alistair E. W. Johnson, Lucas Bulgarelli, and Tom J. Pollard. Deidentification of free-text medical records using pre-trained bidirectional transformers. In Proceedings of the ACM Conference on Health, Inference, and Learning . ACM, apr 2020

2020
[37]

Kohli, Marc B

Dina Demner-Fushman, Marc D. Kohli, Marc B. Rosenman, Sonya E. Shooshan, Laritza Rodriguez, Sameer Antani, George R. Thoma, and Clement J. McDonald. Preparing a col- lection of radiology examinations for distribution and retrieval. Journal of the American Medical Informatics A...

2015
[38]

Annotated dataset creation through general purpose language models for non-english medical nlp, 2022

Johann Frei and Frank Kramer. Annotated dataset creation through general purpose language models for non-english medical nlp, 2022

2022
[39]

Joyce, Niall Taylor, Alejo Nevado-Holgado, Andrea Cipriani, and An- drey Kormilitzin

Isabelle Lorge, Dan W. Joyce, Niall Taylor, Alejo Nevado-Holgado, Andrea Cipriani, and An- drey Kormilitzin. Detecting the clinical features of difficult-to-treat depression using synthetic data from large language models, 2024

2024
[40]

Does synthetic data generation of llms help clinical text mining?, 2023

Ruixiang Tang, Xiaotian Han, Xiaoqian Jiang, and Xia Hu. Does synthetic data generation of llms help clinical text mining?, 2023

2023
[41]

Two directions for clinical data generation with large language models: Data-to-label and label-to-data

Rumeng Li, Xun Wang, and Hong Yu. Two directions for clinical data generation with large language models: Data-to-label and label-to-data. In Findings of the Association for Computational Linguistics: EMNLP 2023 , page 7129–7143. Association for Computational Linguistics, 2023

2023
[42]

An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data, August 2019

Suranga N Kasthurirathne, Gregory Dexter, and Shaun J Grannis. An adversorial approach to enable re-use of machine learning models and collaborative research efforts using synthetic unstructured free-text medical data, August 2019

2019
[43]

Are synthetic clinical notes useful for real natural language processing tasks: A case study on clinical entity recognition

Jianfu Li, Yujia Zhou, Xiaoqian Jiang, Karthik Natarajan, Serguei Vs Pakhomov, Hongfang Liu, and Hua Xu. Are synthetic clinical notes useful for real natural language processing tasks: A case study on clinical entity recognition. Journal of the American Medical Informatics Ass...

2021
[44]

Cardinal, Angus Roberts, Robert Stewart, and Sumithra Velupillai

Julia Ive, Natalia Viani, Joyce Kam, Lucia Yin, Somain Verma, Stephen Puntis, Rudolf N. Cardinal, Angus Roberts, Robert Stewart, and Sumithra Velupillai. Generation and evalua- tion of artificial mental health records for natural language processing. npj Digit. Med. , 3(1), may 2020

2020
[45]

Generative adversarial networks for creating synthetic free-text medical data: A proposal for collaborative research and re-use of machine learning models, May 2021

Suranga N Kasthurirathne, Gregory Dexter, and Shaun J Grannis. Generative adversarial networks for creating synthetic free-text medical data: A proposal for collaborative research and re-use of machine learning models, May 2021

2021
[46]

Zixu Wang, Julia Ive, Sumithra Velupillai, and Lucia Specia. Is artificial data useful for biomedical natural language processing algorithms? In Proceedings of the 18th BioNLP Workshop and Shared Task , pages 240–249, Florence, Italy, aug 2019. Association for Com- putational ...

2019
[47]

Transforming health- care documentation: harnessing the potential of ai to generate discharge summaries

Reece Alexander James Clough, William Anthony Sparkes, Oliver Thomas Clough, Joshua Thomas Sykes, Alexander Thomas Steventon, and Kate King. Transforming health- care documentation: harnessing the potential of ai to generate discharge summaries. BJGP Open, 8(1), 2024

2024
[48]

Transformer models trained on mimic-iii to generate synthetic patient notes, 2020

Ali Amin-Nejad, Julia Ive, and Sumithra Velupillai. Transformer models trained on mimic-iii to generate synthetic patient notes, 2020

2020
[49]

Identifying and aligning medical claims made on social media with medical evidence, 2024

Anthony Hughes and Xingyi Song. Identifying and aligning medical claims made on social media with medical evidence, 2024

2024
[50]

Medically aware gpt-3 as a data generator for medical dialogue summarization, 2021

Bharath Chintagunta, Namit Katariya, Xavier Amatriain, and Anitha Kannan. Medically aware gpt-3 as a data generator for medical dialogue summarization, 2021

2021
[51]

Generaci´ on masiva de historias cl ´ ınicas sint´ eticas con chatgpt: un ejemplo en fractura de cadera

Isidoro Calvo-Lorenzo and Iker Uriarte-Llano. Generaci´ on masiva de historias cl ´ ınicas sint´ eticas con chatgpt: un ejemplo en fractura de cadera. Medicina Cl ´ ınica, 162(11):549– 554, 2024

2024
[52]

Constructing synthetic datasets with generative artificial intelligence to train large language models to classify acute renal failure from clinical notes

Onkar Litake, Brian H Park, Jeffrey L Tully, and Rodney A Gabriel. Constructing synthetic datasets with generative artificial intelligence to train large language models to classify acute renal failure from clinical notes. Journal of the American Medical Informatics Associatio...

2024
[53]

How to use language models for synthetic text generation in cerebrovascular disease-specific medical reports

Byoung-Doo Oh, Gi-Youn Kim, Chulho Kim, and Yu-Seop Kim. How to use language models for synthetic text generation in cerebrovascular disease-specific medical reports. In Ameet Deshpande, EunJeong Hwang, Vishvak Murahari, Joon Sung Park, Diyi Yang, Ashish Sab- harwal, Karthik N...

2024
[54]

Smit h, Nima PourNejatian, Anthony B

Cheng Peng, Xi Yang, Aokun Chen, Kaleb E. Smit h, Nima PourNejatian, Anthony B. Costa, Cheryl Martin, Mona G. Flores, Ying Zhang, Tanja Magoc, Gloria Lipori, Duane A. Mitchell, Naykky S. Ospina, Mustafa M. Ahmed, William R. Hogan, Elizabeth A. Shenkman, Yi Guo, Jiang Bian, and...

2023
[56]

Towards automatic generation of shareable syn- thetic clinical notes using neural language models

Oren Melamud and Chaitanya Shivade. Towards automatic generation of shareable syn- thetic clinical notes using neural language models. In Proceedings of the 2nd Clinical Natural Language Processing Workshop, pages 35–45, Minneapolis, Minnesota, USA, June 2019. As- sociation fo...

2019
[57]

Differentially private medical texts generation using generative neural networks

Md Momin Al Aziz, Tanbir Ahmed, Tasnia Faequa, Xiaoqian Jiang, Yiyu Yao, and Noman Mohammed. Differentially private medical texts generation using generative neural networks. ACM Transactions on Computing for Healthcare , 3(1):1–27, jan 2022

2022
[58]

Nina Zhou, Qiucheng Wu, Zewen Wu, Simeone Marino, and Ivo D. Dinov. DataSifterText: Partially synthetic text generation for sensitive clinical notes. J Med Syst , 46(12), nov 2022. 18

2022
[59]

Knowledge-infused prompting: Assessing and advancing clinical text data generation with large language models

Ran Xu, Hejie Cui, Yue Yu, Xuan Kan, Wenqi Shi, Yuchen Zhuang, May Dongmei Wang, Wei Jin, Joyce Ho, and Carl Yang. Knowledge-infused prompting: Assessing and advancing clinical text data generation with large language models. In Lun-Wei Ku, Andre Martins, and Vivek Srikumar, e...

2024
[60]

Generation and eval- uation of synthetic endoscopy free-text reports with differential privacy

Agathe Zecevic, Xinyue Zhang, Sebastian Zeki, and Angus Roberts. Generation and eval- uation of synthetic endoscopy free-text reports with differential privacy. In Dina Demner- Fushman, Sophia Ananiadou, Makoto Miwa, Kirk Roberts, and Junichi Tsujii, editors, Pro- ceedings of ...

2024
[61]

Generating synthetic documents with clinical keywords: A privacy-sensitive methodology

Simon Meoni, ´Eric De la Clergerie, and Th´ eo Ryffel. Generating synthetic documents with clinical keywords: A privacy-sensitive methodology. In Dina Demner-Fushman, Sophia Ana- niadou, Paul Thompson, and Brian Ondov, editors, Proceedings of the First Workshop on Patient-Orie...

2024
[62]

SynthNotes: A gen- erator framework for high-volume, high-fidelity synthetic mental health notes

Edmon Begoli, Kris Brown, Sudarshan Srinivas, and Suzanne Tamang. SynthNotes: A gen- erator framework for high-volume, high-fidelity synthetic mental health notes. In 2018 IEEE International Conference on Big Data (Big Data) . IEEE, dec 2018

2018
[63]

Aci-bench: a novel ambient clinical intelligence dataset for benchmarking automatic visit note generation

Wen-wai Yim, Yujuan Fu, Asma Ben Abacha, Neal Snider, Thomas Lin, and Meliha Yetisgen. Aci-bench: a novel ambient clinical intelligence dataset for benchmarking automatic visit note generation. Scientific Data , 10(1), September 2023

2023
[64]

An empirical study of clinical note generation from doctor-patient encounters

Asma Ben Abacha, Wen-wai Yim, Yadan Fan, and Thomas Lin. An empirical study of clinical note generation from doctor-patient encounters. In Andreas Vlachos and Isabelle Augenstein, editors, Proceedings of the 17th Conference of the European Chapter of the Association for Comput...

2023
[65]

Feng, Vivek Khetan, Bogdan Sacaleanu, Anatole Gershman, and Eduard Hovy

Steven Y. Feng, Vivek Khetan, Bogdan Sacaleanu, Anatole Gershman, and Eduard Hovy. Chard: Clinical health-aware reasoning across dimensions for text generation models, 2022

2022
[66]

Synth-sbdh: A synthetic dataset of social and behavioral determinants of health for clinical text, 2024

Avijit Mitra, Emily Druhl, Raelene Goodwin, and Hong Yu. Synth-sbdh: A synthetic dataset of social and behavioral determinants of health for clinical text, 2024

2024
[67]

Ehr-ds-qa: A synthetic qa dataset derived from medical dis- charge summaries for enhanced medical information retrieval systems, 2024

Konstantin Kotschenreuther. Ehr-ds-qa: A synthetic qa dataset derived from medical dis- charge summaries for enhanced medical information retrieval systems, 2024

2024
[68]

coherent data set

Jason Walonoski, Dylan Hall, Karen M. Bates, M. Heath Farris, Joseph Dagher, Matthew E. Downs, Ryan T. Sivek, Ben Wellner, Andrew Gregorowicz, Marc Hadley, Francis X. Campion, Lauren Levine, Kevin Wacome, Geoff Emmer, Aaron Kemmer, Maha Malik, Jonah Hughes, Eldesia Granger, an...

2022
[69]

Generating continuous representations of medical texts, 2018

Graham Spinks and Marie-Francine Moens. Generating continuous representations of medical texts, 2018. 19

2018
[70]

On the automatic generation of medical imaging reports

Baoyu Jing, Pengtao Xie, and Eric Xing. On the automatic generation of medical imaging reports. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) . Association for Computational Linguistics, 2018

2018
[71]

Kelly Reynolds, Daniel Nadelman, Joseph Durgin, Stephen Ansah-Addo, Daniel Cole, Rachel Fayne, Jane Harrell, Madison Ratycz, Mason Runge, Amanda Shepard-Hayes, Daniel Wenzel, and Trilokraj Tejasvi. Comparing the quality of chatgpt- and physician-generated responses to patients...

2024
[72]

Vision-language model for generating textual descriptions from clinical images: Model development and validation study

Jia Ji, Yongshuai Hou, Xinyu Chen, Youcheng Pan, and Yang Xiang. Vision-language model for generating textual descriptions from clinical images: Model development and validation study. JMIR Form Res, 8:e32690, Feb 2024

2024
[73]

Generation of natural-language textual summaries from longitudinal clinical records

Goldstein Ayelet and Shahar Yuval. Generation of natural-language textual summaries from longitudinal clinical records. In MEDINFO 2015: eHealth-enabled Health . IOS Press, 2015

2015
[74]

Enhancing clinical note gener- ation from doctor-patient conversations through semantic partition-oriented summarization

Binh-Nguyen Nguyen, Hoang-Quynh Le, and Duy-Cat Can. Enhancing clinical note gener- ation from doctor-patient conversations through semantic partition-oriented summarization. In 2023 15th International Conference on Knowledge and Systems Engineering (KSE) , pages 1–6, 2023

2023
[75]

Enhancing clinical documentation with synthetic data: Leveraging generative models for improved accuracy

Anjanava Biswas and Wrick Talukdar. Enhancing clinical documentation with synthetic data: Leveraging generative models for improved accuracy. International Journal of Innovative Science and Research Technology (IJISRT), page 1553–1566, June 2024

2024
[76]

Medt2t: An adaptive pointer constrain generating method for a new medical text-to-table task

Wang Zhao, Dongxiao Gu, Xuejie Yang, Meihuizi Jia, Changyong Liang, Xiaoyu Wang, and Oleg Zolotarev. Medt2t: An adaptive pointer constrain generating method for a new medical text-to-table task. Future Generation Computer Systems , 161:586–600, 2024

2024
[77]

Li, Xiaodan Liang, Zhiting Hu, and Eric P

Christy Y. Li, Xiaodan Liang, Zhiting Hu, and Eric P. Xing. Knowledge-driven encode, retrieve, paraphrase for medical image report generation, 2019

2019
[78]

Implementation of gpt models for text generation in healthcare domain

Anirban Karak and Kaustuv Kunal. Implementation of gpt models for text generation in healthcare domain. In 2023 International Conference on New Frontiers in Communication, Automation, Management and Security (ICCAMS) , volume 1, pages 1–6, 2023

2023
[79]

Papie˙ z, and Mohammad Yaqub

Santosh Sanjeev, Fadillah Adamsyah Maani, Arsen Abzhanov, Vijay Ram Papineni, Ibrahim Almakky, Bart lomiej W. Papie˙ z, and Mohammad Yaqub. Tibix: Leveraging temporal infor- mation for bidirectional x-ray and report generation, 2024

2024
[80]

Set to ordered text: Generating discharge instruc- tions from medical billing codes

Litton J Kurisinkel and Nancy Chen. Set to ordered text: Generating discharge instruc- tions from medical billing codes. In Kentaro Inui, Jing Jiang, Vincent Ng, and Xiaojun Wan, editors, Proceedings of the 2019 Conference on Empirical Methods in Natural Lan- guage Processing ...

2019
[81]

Visual-textual attentive semantic consistency for medical report generation

Yi Zhou, Lei Huang, Tao Zhou, Huazhu Fu, and Ling Shao. Visual-textual attentive semantic consistency for medical report generation. In 2021 IEEE/CVF International Conference on Computer Vision (ICCV) , pages 3965–3974, 2021. 20

2021
[82]

Fan Bai, Yuxin Du, Tiejun Huang, Max Q. H. Meng, and Bo Zhao. M3d: Advancing 3d medical image analysis with multi-modal large language models, 2024

2024
[83]

Medical scientific table-to-text generation with human-in-the-loop under the data spar- sity constraint, 2022

Heng-Yi Wu, Jingqing Zhang, Julia Ive, Tong Li, Vibhor Gupta, Bingyuan Chen, and Yike Guo. Medical scientific table-to-text generation with human-in-the-loop under the data spar- sity constraint, 2022

2022
[84]

Neural text generation in regulatory medical writing

Claudia Meyer, Daniel Adkins, Koyena Pal, Ruggero Galici, Augusto Garcia-Agundez, and Carsten Eickhoff. Neural text generation in regulatory medical writing. Frontiers in Phar- macology, 14, February 2023

2023
[85]

Prott3: Protein-to-text generation for text-based protein understanding, 2024

Zhiyuan Liu, An Zhang, Hao Fei, Enzhi Zhang, Xiang Wang, Kenji Kawaguchi, and Tat-Seng Chua. Prott3: Protein-to-text generation for text-based protein understanding, 2024

2024
[86]

Multi- modal understanding and generation for medical images and text via vision-language pre- training

Jong Hak Moon, Hyungyung Lee, Woncheol Shin, Young-Hak Kim, and Edward Choi. Multi- modal understanding and generation for medical images and text via vision-language pre- training. IEEE Journal of Biomedical and Health Informatics , 26(12):6070–6080, 2022

2022
[87]

Medm2g: Unifying medical multi-modal generation via cross-guided diffusion with visual invariant, 2024

Chenlu Zhan, Yu Lin, Gaoang Wang, Hongwei Wang, and Jian Wu. Medm2g: Unifying medical multi-modal generation via cross-guided diffusion with visual invariant, 2024

2024
[88]

Improving radiology report generation quality and diversity through reinforcement learning and text augmentation

Daniel Parres, Alberto Albiol, and Roberto Paredes. Improving radiology report generation quality and diversity through reinforcement learning and text augmentation. Bioengineering, 11(4):351, April 2024

2024
[89]

Generat- ing explanations in medical question-answering by expectation maximization inference over evidence, 2023

Wei Sun, Mingxiao Li, Damien Sileo, Jesse Davis, and Marie-Francine Moens. Generat- ing explanations in medical question-answering by expectation maximization inference over evidence, 2023

2023
[90]

Peter J. Liu. Learning to write notes in electronic health records, 2018

2018
[91]

A dictionary-based oversampling approach to clinical document classification on small and imbalanced dataset

Mahdi Abdollahi, Xiaoying Gao, Yi Mei, Shameek Ghosh, and Jinyan Li. A dictionary-based oversampling approach to clinical document classification on small and imbalanced dataset. In 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Tech...

2020
[92]

A simple data augmentation method to im- prove the performance of named entity recognition models in medical domain

Abdul Majeed Issifu and Murat Can Ganiz. A simple data augmentation method to im- prove the performance of named entity recognition models in medical domain. In 2021 6th International Conference on Computer Science and Engineering (UBMK) . IEEE, sep 2021

2021
[93]

An exploratory study on pseudo-data generation in prescription and adverse drug reaction extraction, August 2019

Carson Tao, Kahyun Lee, Michele Filannino, and ¨Ozlem Uzuner. An exploratory study on pseudo-data generation in prescription and adverse drug reaction extraction, August 2019

2019
[94]

An NLP-inspired data augmen- tation method for adverse event prediction using an imbalanced healthcare dataset

Tomoki Ishikawa, Takahiro Yakoh, and Hisashi Urushihara. An NLP-inspired data augmen- tation method for adverse event prediction using an imbalanced healthcare dataset. IEEE Access, 10:81166–81176, 2022

2022
[95]

Exploring transformer text generation for medical dataset augmentation

Ali Amin-Nejad, Julia Ive, and Sumithra Velupillai. Exploring transformer text generation for medical dataset augmentation. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 4699–4708, Marseille, France, May 2020. European Language Resources Ass...

2020
[96]

Substituting clinical features using synthetic medical phrases: Medical text data augmenta- tion techniques

Mahdi Abdollahi, Xiaoying Gao, Yi Mei, Shameek Ghosh, Jinyan Li, and Michael Narag. Substituting clinical features using synthetic medical phrases: Medical text data augmenta- tion techniques. Artificial Intelligence in Medicine , 120:102167, oct 2021

2021
[97]

Textual data augmentation for patient out- comes prediction

Qiuhao Lu, Dejing Dou, and Thien Huu Nguyen. Textual data augmentation for patient out- comes prediction. In 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, dec 2021

2021
[98]

UMLS-based data augmentation for natural language processing of clinical research literature

Tian Kang, Adler Perotte, Youlan Tang, Casey Ta, and Chunhua Weng. UMLS-based data augmentation for natural language processing of clinical research literature. Journal of the American Medical Informatics Association, 28(4):812–823, dec 2020

2020
[99]

Evaluation and analysis of large language models for clinical text augmentation and generation

Atif Latif and Jihie Kim. Evaluation and analysis of large language models for clinical text augmentation and generation. IEEE Access, 12:48987–48996, 2024

2024
[101]

A new generative model for textual descriptions of medical images using transformers enhanced with convolutional neural networks

Artur Gomes Barreto, Juliana Martins de Oliveira, Francisco Nauber Bernardo Gois, Paulo Cesar Cortez, and Victor Hugo Costa de Albuquerque. A new generative model for textual descriptions of medical images using transformers enhanced with convolutional neural networks. Bioengi...

2023
[102]

Rundensteiner

ML Tlachac, Walter Gerych, Kratika Agrawal, Benjamin Litterer, Nicholas Jurovich, Saitheeraj Thatigotla, Jidapa Thadajarassiri, and Elke A. Rundensteiner. Text generation to aid depression detection: A comparative study of conditional sequence generative adversarial networks. ...

2022
[103]

Generating and evaluating simulated medical notes: Getting a natural language generation model to give you what you want

Robert Horton, Maryam Hosseinabadi, Alexandre Vilcek, Wolfgang Pauli, and Mario In- chiosa. Generating and evaluating simulated medical notes: Getting a natural language generation model to give you what you want. In Document Intelligence Workshop at KDD, 2021, Virtual Event ,...

2021
[104]

Relation extraction in underex- plored biomedical domains: A diversity-optimized sampling and synthetic data generation approach

Maxime Delmas, Magdalena Wysocka, and Andr´ e Freitas. Relation extraction in underex- plored biomedical domains: A diversity-optimized sampling and synthetic data generation approach. Computational Linguistics , 50(3):953–1000, 09 2024

2024
[105]

Generation and de-identification of indian clinical discharge summaries using llms, 2024

Sanjeet Singh, Shreya Gupta, Niralee Gupta, Naimish Sharma, Lokesh Srivastava, Vibhu Agarwal, and Ashutosh Modi. Generation and de-identification of indian clinical discharge summaries using llms, 2024

2024
[106]

Liu, Vijay Prakash Dwivedi, Thanh-Tung Nguyen, Xiaoxue Gao, Nancy F

Kuluhan Binici, Abhinav Ramesh Kashyap, Viktor Schlegel, Andy T. Liu, Vijay Prakash Dwivedi, Thanh-Tung Nguyen, Xiaoxue Gao, Nancy F. Chen, and Stefan Winkler. Medsage: Enhancing robustness of medical dialogue summarization to asr errors with llm-generated synthetic dialogues, 2024

2024
[107]

Pulsar at mediqa-sum 2023: Large language models aug- mented by synthetic dialogue convert patient dialogues to medical records

Viktor Schlegel, Hao Li, Yuping Wu, Anand Subramanian, Thanh-Tung Nguyen, Abhi- nav Ramesh Kashyap, Daniel Beck, Xiaojun Zeng, Riza Theresa Batista-Navarro, Stefan Winkler, and Goran Nenadic. Pulsar at mediqa-sum 2023: Large language models aug- mented by synthetic dialogue co...

2023
[108]

Jihye Kim Scroggins, Maxim Topaz, Jiyoun Song, and Maryam Zolnoori. Does synthetic data augmentation improve the performances of machine learning classifiers for identifying health problems in patient–nurse verbal communications in home healthcare settings? Journal of Nursing ...

2024
[109]

Use of large language models for medical synthetic data generation in mental illness

˙Irfan AYG¨UN and Mehmet KAYA. Use of large language models for medical synthetic data generation in mental illness. In 7th IET Smart Cities Symposium (SCS 2023) , volume 2023, pages 652–656, 2023

2023
[110]

ECG-QALM: Entity-controlled synthetic text generation using contextual Q&A for NER

Karan Aggarwal, Henry Jin, and Aitzaz Ahmad. ECG-QALM: Entity-controlled synthetic text generation using contextual Q&A for NER. In Anna Rogers, Jordan Boyd-Graber, and Naoaki Okazaki, editors, Findings of the Association for Computational Linguistics: ACL 2023 , pages 5649–56...

2023
[111]

Jorre, Claudia Pagliari, Ruth Jepson, and Sarah Cunningham- Burley

Mhairi Aitken, Jenna de St. Jorre, Claudia Pagliari, Ruth Jepson, and Sarah Cunningham- Burley. Public responses to the sharing and linkage of health data for research purposes: a systematic review and thematic synthesis of qualitative studies. BMC Medical Ethics, 17(1), November 2016

2016
[112]

giving something back

Jessica Stockdale, Jackie Cassell, and Elizabeth Ford. “giving something back”: A systematic review and ethical enquiry into public views on the use of patient data for research in the united kingdom and the republic of ireland. Wellcome Open Research, 3:6, January 2019

2019
[113]

Multilayer perceptron: Architecture optimization and training

Hassan Ramchoun, Mohammed Amine, Janati Idrissi, Youssef Ghanou, and Mohamed Et- taouil. Multilayer perceptron: Architecture optimization and training. International Journal of Interactive Multimedia and Artificial Intelligence , 4(1):26, 2016

2016
[114]

word2vec explained: deriving mikolov et al.’s negative- sampling word-embedding method, 2014

Yoav Goldberg and Omer Levy. word2vec explained: deriving mikolov et al.’s negative- sampling word-embedding method, 2014

2014
[115]

BioWordVec, im- proving biomedical word embeddings with subword information and MeSH

Yijia Zhang, Qingyu Chen, Zhihao Yang, Hongfei Lin, and Zhiyong Lu. BioWordVec, im- proving biomedical word embeddings with subword information and MeSH. Scientific Data , 6(1), May 2019

2019
[116]

Recurrent neural network regulariza- tion, 2014

Wojciech Zaremba, Ilya Sutskever, and Oriol Vinyals. Recurrent neural network regulariza- tion, 2014

2014
[117]

Long Short-Term Memory , pages 37–45

Alex Graves. Long Short-Term Memory , pages 37–45. Springer Berlin Heidelberg, Berlin, Heidelberg, 2012

2012
[118]

Bidirectional re- current neural network language models for automatic speech recognition

Ebru Arisoy, Abhinav Sethy, Bhuvana Ramabhadran, and Stanley Chen. Bidirectional re- current neural network language models for automatic speech recognition. In 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5421– 5425, 2015

2015
[119]

Empirical evalu- ation of gated recurrent neural networks on sequence modeling, 2014

Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio. Empirical evalu- ation of gated recurrent neural networks on sequence modeling, 2014

2014
[120]

Minh-Thang Luong, Hieu Pham, and Christopher D. Manning. Effective approaches to attention-based neural machine translation, 2015. 23

2015
[121]

Liu, and Christopher D

Abigail See, Peter J. Liu, and Christopher D. Manning. Get to the point: Summarization with pointer-generator networks, 2017

2017
[122]

Auto-encoding variational bayes, 2013

Diederik P Kingma and Max Welling. Auto-encoding variational bayes, 2013

2013
[123]

Deep feature consistent variational autoencoder

Xianxu Hou, Linlin Shen, Ke Sun, and Guoping Qiu. Deep feature consistent variational autoencoder. In 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), pages 1133–1141, 2017

2017
[124]

An introduction to convolutional neural networks, 2015

Keiron O’Shea and Ryan Nash. An introduction to convolutional neural networks, 2015

2015
[125]

High-resolution image synthesis with latent diffusion models, 2022

Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨ orn Ommer. High-resolution image synthesis with latent diffusion models, 2022

2022
[126]

Varshney, Caiming Xiong, and Richard Socher

Nitish Shirish Keskar, Bryan McCann, Lav R. Varshney, Caiming Xiong, and Richard Socher. Ctrl: A conditional transformer language model for controllable generation, 2019

2019
[127]

Supervision exists everywhere: A data efficient contrastive language-image pre-training paradigm, 2022

Yangguang Li, Feng Liang, Lichen Zhao, Yufeng Cui, Wanli Ouyang, Jing Shao, Fengwei Yu, and Junjie Yan. Supervision exists everywhere: A data efficient contrastive language-image pre-training paradigm, 2022

2022
[128]

Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks, 2019

Jiasen Lu, Dhruv Batra, Devi Parikh, and Stefan Lee. Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks, 2019

2019
[129]

Seongjun Yun, Minbyul Jeong, Raehyun Kim, Jaewoo Kang, and Hyunwoo J. Kim. Graph transformer networks, 2020

2020
[130]

Peters, and Arman Cohan

Iz Beltagy, Matthew E. Peters, and Arman Cohan. Longformer: The long-document trans- former, 2020

2020
[131]

Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension, 2019

Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Ves Stoyanov, and Luke Zettlemoyer. Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension, 2019

2019
[132]

Biobart: Pretraining and evaluation of a biomedical generative language model, 2022

Hongyi Yuan, Zheng Yuan, Ruyi Gan, Jiaxing Zhang, Yutao Xie, and Sheng Yu. Biobart: Pretraining and evaluation of a biomedical generative language model, 2022

2022
[133]

Eduardo Ares, and Javier Parapar

Jorge Gab ´ ın, M. Eduardo Ares, and Javier Parapar. Enhancing automatic keyphrase la- belling with text-to-text transfer transformer (t5) architecture: A framework for keyphrase generation and filtering, 2024

2024
[134]

Gen- erating medical prescriptions with conditional transformer, 2023

Samuel Belkadi, Nicolo Micheletti, Lifeng Han, Warren Del-Pinto, and Goran Nenadic. Gen- erating medical prescriptions with conditional transformer, 2023

2023
[135]

Large language models generate functional protein sequences across diverse families

Ali Madani, Ben Krause, Eric R Greene, Subu Subramanian, Benjamin P Mohr, James M Holton, Jose Luis Olmos Jr, Caiming Xiong, Zachary Z Sun, Richard Socher, et al. Large language models generate functional protein sequences across diverse families. Nature biotech- nology, 41(8)...

2023
[136]

Supervision exists everywhere: A data efficient contrastive language-image pre-training paradigm, 2021

Yangguang Li, Feng Liang, Lichen Zhao, Yufeng Cui, Wanli Ouyang, Jing Shao, Fengwei Yu, and Junjie Yan. Supervision exists everywhere: A data efficient contrastive language-image pre-training paradigm, 2021. 24

2021
[137]

Blip: Bootstrapping language- image pre-training for unified vision-language understanding and generation

Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. Blip: Bootstrapping language- image pre-training for unified vision-language understanding and generation. In International conference on machine learning , pages 12888–12900. PMLR, 2022

2022
[138]

Vision-language model for generating textual descriptions from clinical images: Model development and validation study

Jia Ji, Yongshuai Hou, Xinyu Chen, Youcheng Pan, and Yang Xiang. Vision-language model for generating textual descriptions from clinical images: Model development and validation study. JMIR Formative Research, 8:e32690, February 2024

2024
[139]

Bert: Pre-training of deep bidirectional transformers for language understanding, 2018

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding, 2018

2018
[140]

Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter, 2019

Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter, 2019

2019
[141]

A pubmedbert-based classifier with data aug- mentation strategy for detecting medication mentions in tweets, 2021

Qing Han, Shubo Tian, and Jinfeng Zhang. A pubmedbert-based classifier with data aug- mentation strategy for detecting medication mentions in tweets, 2021

2021
[142]

Smed- bert: A knowledge-enhanced pre-trained language model with structured semantics for med- ical text mining, 2021

Taolin Zhang, Zerui Cai, Chengyu Wang, Minghui Qiu, Bite Yang, and Xiaofeng He. Smed- bert: A knowledge-enhanced pre-trained language model with structured semantics for med- ical text mining, 2021

2021
[143]

A short study on compressing decoder-based language models, 2021

Tianda Li, Yassir El Mesbahi, Ivan Kobyzev, Ahmad Rashid, Atif Mahmud, Nithin Anchuri, Habib Hajimolahoseini, Yang Liu, and Mehdi Rezagholizadeh. A short study on compressing decoder-based language models, 2021

2021
[144]

Not enough data? deep learning to the rescue!, 2019

Ateret Anaby-Tavor, Boaz Carmeli, Esther Goldbraich, Amir Kantor, George Kour, Segev Shlomov, Naama Tepper, and Naama Zwerdling. Not enough data? deep learning to the rescue!, 2019

2019
[145]

Liu, Mohammad Saleh, Etienne Pot, Ben Goodrich, Ryan Sepassi, Lukasz Kaiser, and Noam Shazeer

Peter J. Liu, Mohammad Saleh, Etienne Pot, Ben Goodrich, Ryan Sepassi, Lukasz Kaiser, and Noam Shazeer. Generating wikipedia by summarizing long sequences, 2018

2018
[146]

Biogpt: generative pre-trained transformer for biomedical text generation and mining

Renqian Luo, Liai Sun, Yingce Xia, Tao Qin, Sheng Zhang, Hoifung Poon, and Tie-Yan Liu. Biogpt: generative pre-trained transformer for biomedical text generation and mining. Briefings in Bioinformatics , 23(6):bbac409, 09 2022

2022
[147]

Galactica: A large language model for science, 2022

Ross Taylor, Marcin Kardas, Guillem Cucurull, Thomas Scialom, Anthony Hartshorn, Elvis Saravia, Andrew Poulton, Viktor Kerkez, and Robert Stojnic. Galactica: A large language model for science, 2022

2022
[148]

Gpt- neox-20b: An open-source autoregressive language model, 2022

Sid Black, Stella Biderman, Eric Hallahan, Quentin Anthony, Leo Gao, Laurence Golding, Ho- race He, Connor Leahy, Kyle McDonell, Jason Phang, Michael Pieler, USVSN Sai Prashanth, Shivanshu Purohit, Laria Reynolds, Jonathan Tow, Ben Wang, and Samuel Weinbach. Gpt- neox-20b: An ...

2022
[149]

How to use language models for synthetic text generation in cerebrovascular disease-specific medical reports

Byoung-Doo Oh, Gi-Youn Kim, Chulho Kim, and Yu-Seop Kim. How to use language models for synthetic text generation in cerebrovascular disease-specific medical reports. InProceedings of the 1st Workshop on Personalization of Generative AI Systems (PERSONALIZE 2024) , pages 10–17, 2024

2024
[150]

Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks, 2014. 25

2014
[151]

Seqgan: Sequence generative adversarial nets with policy gradient, 2016

Lantao Yu, Weinan Zhang, Jun Wang, and Yong Yu. Seqgan: Sequence generative adversarial nets with policy gradient, 2016

2016
[152]

Deep reinforcement learning: An overview, 2017

Yuxi Li. Deep reinforcement learning: An overview, 2017

2017
[153]

Adversarial sub-sequence for text generation, 2019

Xingyuan Chen, Yanzhe Li, Peng Jin, Jiuhua Zhang, Xinyu Dai, Jiajun Chen, and Gang Song. Adversarial sub-sequence for text generation, 2019

2019
[154]

Rush, and Yann LeCun

Jake Zhao, Yoon Kim, Kelly Zhang, Alexander M. Rush, and Yann LeCun. Adversarially regularized autoencoders, 2017

2017
[155]

Bleu: a method for auto- matic evaluation of machine translation

Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. Bleu: a method for auto- matic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics , pages 311–318, Philadelphia, Pennsylvania, USA, July 2002...

2002
[156]

On accurate evaluation of gans for language generation, 2018

Stanislau Semeniuta, Aliaksei Severyn, and Sylvain Gelly. On accurate evaluation of gans for language generation, 2018

2018
[157]

Texygen: A benchmarking platform for text generation models, 2018

Yaoming Zhu, Sidi Lu, Lei Zheng, Jiaxian Guo, Weinan Zhang, Jun Wang, and Yong Yu. Texygen: A benchmarking platform for text generation models, 2018

2018
[158]

Yonghui Wu, Mike Schuster, Zhifeng Chen, Quoc V. Le, Mohammad Norouzi, Wolfgang Macherey, Maxim Krikun, Yuan Cao, Qin Gao, Klaus Macherey, Jeff Klingner, Apurva Shah, Melvin Johnson, Xiaobing Liu, Lukasz Kaiser, Stephan Gouws, Yoshikiyo Kato, Taku Kudo, Hideto Kazawa, Keith St...

2016
[159]

ROUGE: A package for automatic evaluation of summaries

Chin-Yew Lin. ROUGE: A package for automatic evaluation of summaries. In Text Summa- rization Branches Out , pages 74–81, Barcelona, Spain, July 2004. Association for Computa- tional Linguistics

2004
[160]

Meteor universal: Language specific translation evalu- ation for any target language

Michael Denkowski and Alon Lavie. Meteor universal: Language specific translation evalu- ation for any target language. In Proceedings of the Ninth Workshop on Statistical Machine Translation, pages 376–380, Baltimore, Maryland, USA, June 2014. Association for Compu- tational ...

2014
[161]

CharacTer: Translation edit rate on character level

Weiyue Wang, Jan-Thorsten Peter, Hendrik Rosendahl, and Hermann Ney. CharacTer: Translation edit rate on character level. In Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers , pages 505–510, Berlin, Germany, August 2016. Association for ...

2016
[162]

A study of translation edit rate with targeted human annotation

Matthew Snover, Bonnie Dorr, Rich Schwartz, Linnea Micciulla, and John Makhoul. A study of translation edit rate with targeted human annotation. In Proceedings of the 7th Conference of the Association for Machine Translation in the Americas: Technical Papers , pages 223– 231, ...

2006
[163]

Lawrence Zitnick, and Devi Parikh

Ramakrishna Vedantam, C. Lawrence Zitnick, and Devi Parikh. Cider: Consensus-based image description evaluation, 2014. 26

2014
[164]

Understanding inverse document frequency: on theoretical arguments for IDF

Stephen Robertson. Understanding inverse document frequency: on theoretical arguments for IDF. Journal of Documentation , 60(5):503–520, October 2004

2004
[165]

The probabilistic relevance framework: Bm25 and beyond

Stephen Robertson, Hugo Zaragoza, et al. The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends® in Information Retrieval , 3(4):333–389, 2009

2009
[166]

Bertscore: Evaluating text generation with bert

Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q Weinberger, and Yoav Artzi. Bertscore: Evaluating text generation with bert. arXiv preprint arXiv:1904.09675 , 2019

1904 arXiv
[167]

Semscore: Automated evaluation of instruction-tuned llms based on semantic textual similarity

Ansar Aynetdinov and Alan Akbik. Semscore: Automated evaluation of instruction-tuned llms based on semantic textual similarity. arXiv preprint arXiv:2401.17072 , 2024

2024 arXiv
[168]

Improving radiology report generation quality and diversity through reinforcement learning and text augmentation

Daniel Parres, Alberto Albiol, and Roberto Paredes. Improving radiology report generation quality and diversity through reinforcement learning and text augmentation. Bioengineering, 11(4):351, 2024

2024
[169]

Spice: Semantic propositional image caption evaluation

Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. Spice: Semantic propositional image caption evaluation. In European conference on computer vision , pages 382–398. Springer, 2016

2016
[170]

Marry L. McHugh. Interrater reliability: the kappa statistic. Biochemia Medica, pages 276– 282, 2012

2012
[171]

Extending the cochran rule for the com- parison of word frequencies between corpora

Paul Rayson, Damon Berridge, and Brian Francis. Extending the cochran rule for the com- parison of word frequencies between corpora. In 7th International Conference on Statistical analysis of textual data (JADT 2004) , pages 926–936, 2004

2004
[172]

Speech and language processing (second edition) daniel jurafsky and james h

Vlado Keselj. Speech and language processing (second edition) daniel jurafsky and james h. martin (stanford university and university of colorado at boulder) pearson prentice hall, 2009, xxxi+988 pp; hardbound, ISBN 978-0-13-187321-6, $115.00. Computational Linguistics, 35(3):...

2009
[173]

Yunhui Long, Vincent Bindschaedler, and Carl A. Gunter. Towards measuring membership privacy, 2017

2017
[174]

Thibault Sellam, Dipanjan Das, and Ankur P. Parikh. Bleurt: Learning robust metrics for text generation, 2020. 27 A Appendix A: Generation Purposes A.1 Privacy-Preserving In the medical field, privacy plays a vital role, in which preserving individual information is an essenti...

2020
[176]

2015 Assistive W riting Knowledge Source, T ext Processing T emplate-based Publicly Avail- able English Human Assessment Structure T ext Summarisa- tion Discharge Sum- maries

2015
[177]

2018 Assistive W riting Neural Network ARAE Publicly Avail- able (IU X-RA Y) English PPL Structure, T est on Utility Report Genera- tion Radiology

2018
[178]

2018 Privacy- preserving Neural Network SeqGAN (LSTM as gener- ator, CNN and Bi-LSTM as discriminator), RL Private EHR/EMR Chinese BLEU, NLL, AdvSuc, Hu- man Assessment (ERE) Similarity , Struc- ture, T est on Utility Disease Classifi- cation Discharge Sum- maries

2018
[179]

2018 Assistive W riting Neural Network Seq2Seq (CNN Encoder and LSTM Decoder) Publicly Avail- able (IU X-RA Y) English BLEU, ROUGE, CIDEr, METEOR Similarity , T est on Utility Report Genera- tion Radiology

2018
[180]

2018 Assistive W riting Neural Network T ransformers (TDMCA) Publicly Avail- able (MIMIC-III) English PPL, ROUGE, Human As- sessment Similarity , Struc- ture, T est on Utility Report Genera- tion Discharge Sum- maries

2018
[181]

2018 Corpus Building T ext Process- ing, Knowledge Source UML, T emplate-based Publicly Avail- able (MIMIC-III) English BLEU, METEOR Similarity - Discharge Sum- maries

2018
[182]

2018 Corpus Building T ext Processing T emplate-based Online source German Human Assessment Structure - Clinical Practice Guidelines

2018
[183]

2018 Privacy- preserving, Augmentation Neural Network Seq2Seq (LSTM Encoder and LSTM Decoder) Private EHR/EMR English BLEU, ROUGE, Odd- ratio, CIDEr, Human Assessment (Check Sen- sitive Information), Classification Accuracy Similarity , Struc- ture, Privacy , T est on Utility ...

2018
[184]

2018 Corpus Building, Annotation Manual - Private EHR/EMR Norwagian Human Assessment (IAA), Pattern Match, Classifica- tion Accuracy Structure, T est on Utility Named Entity Recognition, Re- lation Extraction History of Present Illness (HPI)

2018
[185]

2019 Usefulness Neural Network T ransformers (CTRL), RAKE Publicly Avail- able (MIMIC-III) English BLEU, ROUGE Similarity , T est on Utility Phenotype Clas- sification, Rela- tion Extraction Discharge Sum- maries

2019
[186]

2019 Privacy- preserving Neural Network LSTM Publicly Avail- able (MIMIC-III) English PPL, PDTP , Human As- sessment, NLI Similarity , Struc- ture, Privacy - Discharge Sum- maries

2019
[187]

2019 Usefulness, Privacy- preserving Neural Network SeqGAN Private EHR/EMR English BLEU, Hamming Distance Similarity , Pri- vacy , T est on Utility Disease Classifi- cation Patient Labora- tory T est

2019
[188]

2019 Augmentation Knowledge Source Gazetteers Publicly Avail- able English Classification Accuracy T est on Utility Adverse Drug Re- action/Event Ex- traction Drug and Medica- tion

2019
[189]

2019 Assistive W riting Neural Network Seq2Seq (RNN Encoder and RNN Decoder) Publicly Avail- able (MIMIC-III) English BLEU, Human Assessment Similarly , Struc- ture Report Genera- tion Discharge Sum- maries

2019
[190]

2019 Assistive W riting Neural Network GPT-2 Private EHR/EMR Chinese Human Assessment Similarity Report Genera- tion Discharge Sum- maries

2019
[191]

2019 Assistive W riting Neural Network GTR, CNN (DenseNet) Publicly Avail- able (IU X-RA Y) English BLEU, ROUGE, CIDEr, Human Assessment, Clas- sification Accuracy Similarity , Struc- ture, T est on Utility Disease Classifi- cation Radiology Article Y ear Purpose Generation Me...

2019
[192]

2020 Usefulness Neural Network T ransformers (Encoder- Decoder), GPT-2 (De- coder) Publicly Avail- able (MIMIC-III) English Classification Accuracy T est on Utility Phenotype Classification, Readmission Prediction Discharge Sum- maries

2020
[193]

2020 Corpus Building Knowledge Source, T ext Processing UML, SpaCy Online source German Human Assessment (IAA), Pattern Match Structure - Clinical Practice Guidelines

2020
[194]

2020 Assistive W riting, Augmentation Neural Network Seq2Seq (GRU Encoder and GRU Decoder), RL, BER T Publicly Avail- able (MIMIC- CXR) Japanese, English BLEU, ROUGE, Human Assessment Similarity , Struc- ture, T est on Utility Report Genera- tion Radiology

2020
[195]

2020 Corpus Building Manual - Private EHR/EMR Bulgarian Classification Accuracy T est on Utility Phenotype Clas- sification Discharge Sum- maries

2020
[196]

2020 Augmentation Knowledge Source, Neu- ral Network W ordNet, GloV e, word2vec, BioW ord2V ec Publicly Avail- able English Classification Accuracy T est on Utility Disease Classifi- cation Discharge Sum- maries

2020
[197]

2020 Augmentation Neural Network T ransformers, GPT-2 Publicly Avail- able (MIMIC-III) English PPL, BLEU, ROUGE Similarity , T est on Utility Readmission Prediction, Phenotype Clas- sification Discharge Sum- maries

2020
[198]

2020 Privacy- preserving, Usefulness Neural Network T ransformers (CTRL) Private EHR/EMR English BLEU, PPL, TER, Human Assessment (IAA), Classi- fication Accuracy Similarity , Struc- ture, Privacy , T est on Utility Phenotype Clas- sification Discharge Sum- maries

2020
[199]

2021 Augmentation T ext Process- ing, Knowledge Source UML, EDA Publicly Avail- able English Classification Accuracy T est on Utility Named Entity Recognition Population, Intervention, Comparison, and Outcome (PICO)

2021
[200]

2021 Assistive W riting Neural Network Seq2Seq (DenseNet CNN Encoder and LSTM De- coder) Publicly Avail- able (IU X-RA Y, MIMIC-CXR) English BLEU, ROUGE, CIDEr, METEOR, Hamming Dis- tance Similarity , T est on Utility Report Genera- tion Radiology

2021
[201]

2021 Privacy- preserving, Corpus Building Manual crowdsourcing human-in- the-loop Private EHR/EMR Japanese Human Assessment Similarity , Struc- ture, Privacy - Patient Labora- tory T est

2021
[202]

2021 Augmentation, Usefulness Neural Network GPT-3 Manual Collec- tion and Curation English ROUGE, Negation Cor- rectness, Human Assess- ment Similarity , Struc- ture, T est on Utility T ext Summarisa- tion Medical Conver- sations

2021
[203]

2021 Privacy- preserving Neural Network SeqGAN (LSTM as gener- ator, CNN and Bi-LSTM as discriminator), RL Private EHR/EMR Chinese BLEU, AdvSuc, NLL, Hu- man Assessment (ERE) Similarity , Struc- ture, T est on Utility Disease Classifi- cation Discharge Sum- maries

2021
[204]

2021 Usefulness Neural Network T ransformers (CTRL), GPT-2, CharRNN, Seq- GAN Publicly Avail- able English BLEU Similarity , Struc- ture, T est on Utility Named Entity Recognition History of Present Illness (HPI)

2021
[205]

2021 Augmentation Knowledge Source GloV e, W ordNet, UMLs Publicly Avail- able English Classification Accuracy T est on Utility Disease Classifi- cation Discharge Sum- maries

2021
[206]

2021 Augmentation Manual - Private EHR/EMR Norwegian Classification Accuracy T est on Utility Named Entity Recognition, Re- lation Extraction History of Present Illness (HPI)

2021
[207]

2021 Annotation, Pri- vacy Neural Network LSTM, GPT-2 Private EHR/EMR Dutch ROUGE, BM25, Human Assessment (IAA), Pat- tern Match Similarity , Struc- ture, T est on Utility De-Identification Discharge Sum- maries Article Y ear Purpose Generation Method Approach / Architecture D...

2021
[208]

2021 Usefulness, Privacy- preserving Neural Network SeqGAN Private EHR/EMR English BLEU, GLEU, Hamming Distance Similarity , Pri- vacy , T est on Utility Disease Classifi- cation Patient Labora- tory T est

2021
[209]

2021 Augmentation Neural Network distil-GPT-2, LAMBADA Publicly Avail- able (MIMIC-III) English Classification Accuracy T est on Utility Readmission Pre- diction Discharge Sum- maries

2021
[210]

2021 Augmentation T ext Process- ing, Knowledge Source UML, EDA Publicly Avail- able English Classification Accuracy T est on Utility Named Entity Recognition Biological Con- cepts and Rela- tions

2021
[211]

2021 Privacy- preserving Neural Network DP-GPT Publicly Avail- able (MIMIC-III) English NLL, BLEU, Jaccard, Ad- vSuc, DTP Similarity , Struc- ture, Privacy , T est on Utility Disease Classifi- cation Discharge Sum- maries [103] 2021 Augmentation Neural Network T ransformers On...

2021
[212]

2022 Augmentation Knowledge Source Gazetteers Online source Indonesian Classification Accuracy T est on Utility T ext Classifica- tion COVID News

2022
[213]

2022 Annotation, Cor- pus Building Neural Network GPT-neox Prompting German Pattern Match, Classifica- tion Accuracy T est on Utility Named Entity Recognition Drug and Medica- tion [102] 2022 Augmentation Neural Network SeqGAN Publicly Avail- able English NLL, BLEU, Classifica...

2022
[214]

2022 Assistive W riting Neural Network Bi-LSTM Publicly Avail- able Chinese Classification Accuracy Similarity Question Answer- ing Medical Consul- tation

2022
[215]

2022 Assistive W riting Neural Network T ransformers (Encoder- Decoder), SMedBER T Publicly Avail- able Chinese BLEU, Human Assessment Similarity , Struc- ture, T est on Utility Question Answer- ing Medical Conver- sations

2022
[216]

2022 Augmentation, Assistive W riting Neural Network T ransformers (T5) Publicly Avail- able English ROUGE, BLEU, TER, Human Assessment Similarity , Struc- ture, T est on Utility Report Genera- tion Patient Labora- tory T est

2022
[217]

2022 Assistive W riting Neural Network VLP (CNN ResNet, BER T) Publicly Avail- able (MIMIC- CXR) English BLEU, Classification Ac- curacy Similarity , T est on Utility Disease Classi- fication, Report Generation, Question Answer- ing Radiology

2022
[218]

2022 Corpus Building Neural Network GPT-2 Publicly Avail- able (MIMIC-III) English Human Assessment Structure nan Discharge Sum- maries (Cardio- vascular)

2022
[219]

2022 Corpus Building Knowledge Source, T ext Processing UML, SpaCy Online source German Clustering Structure nan Clinical Practice Guidelines

2022
[220]

2022 Augmentation T ext Processing EDA Manual Collec- tion and Curation Japanese Classification Accuracy T est on Utility Phenotype Clas- sification Doctor-Patient Conversations

2022
[221]

2022 Augmentation Neural Network word2vec, Cosine Similar- ity Private EHR/EMR English Classification Accuracy T est on Utility Adverse Drug Re- action/Event Ex- traction Drug and Medica- tion

2022
[222]

2022 Privacy- preserving Knowledge Source, T ext Processing UML, RAKE Publicly Avail- able (MIMIC-III) English Clustering, BLEU, TFIDF, Human As- sessment (T uring T est) Similarity , Struc- ture, Privacy nan Discharge Sum- maries Article Y ear Purpose Generation Method Approa...

2022
[223]

2022 Corpus Building Neural Network T ransformers (T5 and BAR T) Manual Collec- tion and Curation English BLEU, ROUGE-L, CIDEr, METEOR, PPL, SPICE, BER Tscore, Human As- sessment Similarity , Struc- ture - History of Present Illness (HPI) [100] 2023 Augmentation Neural Network...

2022
[224]

2023 Corpus Building Knowledge Source, Neu- ral Network UML, BioBAR T, LED Manual Collec- tion and Curation English ROUGE, BER Tscore, BLEUR T, MedCon Similarity - Doctor-Patient Conversations [101] 2023 Augmentation Neural Network T ransformers, CNN- ResNet Publicly Avail- ab...

2023
[225]

2023 Assistive W riting Neural Network, T ext Processing BAR T, distilBER T, T emplate-based Publicly Avail- able English ROUGE Similarity Question Answer- ing, T ext Summa- rization Doctor-Patient Conversations

2023
[226]

2023 Assistive W riting Neural Network GPT-2 Publicly Avail- able English BLEU, ROUGE Similarity Question Answer- ing Drug and Medica- tion [110] 2023 Augmentation Neural Network GPT-2 Publicly Avail- able English ROUGE-L, PPL Similarity , Struc- ture Named Entity Recognition ...

2023
[227]

2023 Assistive W riting Neural Network Seq2Seq (RNN Encoder and RNN Decoder) Point Generator Network Online source English ROUGE, ROUGE-L, Co- sine, Jaccard, TFIDF Similarity T ext Summarisa- tion Drug and Medica- tion

2023
[228]

2023 Corpus Building, Augmentation Neural Network BAR T Private EHR/EMR English ROUGE, BLEUR T, BER TScore, Human Assessment Similarity , Struc- ture T ext Summarisa- tion Doctor-Patient Conversations

2023
[229]

2023 Augmentation, Annotation, Usefulness Neural Network ChatGPT Prompting English Classification Accuracy , Human Assessment, Pattern Match Similarity , Struc- ture, T est on Utility Named Entity Recognition, Re- lation Extraction Biological Con- cepts and Rela- tions [107] 2...

2023
[230]

2023 Usefulness Neural Network GPT-2 Publicly Avail- able F rench BLEU, Pattern Match, Human Assessment Similarity , Struc- ture, T est on Utility Named Entity Recognition Biological Con- cepts and Rela- tions

2023
[231]

2023 Usefulness Neural Network GPT-3 Private EHR/EMR English Human Assessment (T ur- ing T est), Classification Accuracy Structure, T est on Utility De-Identification, Relation Extrac- tion, Question Answering History of Present Illness (HPI) [109] 2023 Augmentation Neural Net...

2023
[232]

(MIMIC-III) English Human Assessment (Pat- tern Match), Classification Accuracy Structure, T est on Utility Phenotype Clas- sification History of Present Illness (HPI)

2023 Augmentation, Annotation Neural Network GPT-4 Manual Col- lection and Curation, Pub- lic Available. (MIMIC-III) English Human Assessment (Pat- tern Match), Classification Accuracy Structure, T est on Utility Phenotype Clas- sification History of Present Illness (HPI)

2023
[233]

2023 Augmentation, Privacy- preserving Neural Network ChatGPT, PubMedBER T Prompting English Classification Accuracy T est on Utility Relation Extrac- tion, Named En- tity Recognition, Question Answer- ing Clinical T ran- scripts Article Y ear Purpose Generation Method Approac...

2023
[234]

2024 Assistive W riting Neural Network ChatGPT Prompting English Human Assessment Structure, T est on Utility Question Answer- ing Medical Consul- tation

2024
[235]

2024 Assistive W riting Neural Network ClinicalBLIP Publicly Avail- able (IU X-RA Y, MIMIC-CXR) English ROUGE, METEOR Simlarity Report Genera- tion X-Ray

2024
[236]

2024 Augmentation Neural Network, T ext Processing ChatGPT, BAR T, T5, EDA Publicly Avail- able English ROUGE, CIDEr, ME- TEOR, BER Tscore Similarity Diagnose Pre- diction, Disease Classification Population, Intervention, Comparison, and Outcome (PICO)

2024
[237]

Synthetic text) Report Genera- tion Discharge Sum- maries

2024 Usefulness Neural Network ChatGPT Prompting English Human Assessment Structure Com- parison (Human vs. Synthetic text) Report Genera- tion Discharge Sum- maries

2024
[238]

2023 Assistive W riting Neural Network GPT Online source Chinese PPL Structure, T est on Utility Question Answer- ing Doctor-Patient Conversations

2023
[239]

2024 Assistive W riting Neural Network V AE, GAN (LSTM Gener- ator, CNN Discriminator) Publicly Avail- able (MIMIC-III) English BLEU, PPL, WER, Hu- man Assessment Structure, Simi- larity - Clinical T ran- scripts

2024
[240]

2024 Assistive W riting Neural Network BAR T Publicly Avail- able English ROUGE-L, BER Tscore Similarity Report Genera- tion Patient Labora- tory T est [104] 2024 Augmentation Neural Network BioGPT Publicly Avail- able English BLEU, Classification Ac- curacy Similarity , T est...

2024
[241]

2024 Assistive W rit- ing, Privacy- preserving Neural Network DP-GPT, BioGPT Private EHR/EMR English Human Assessment, ROUGE-L, Privacy Bud- get Similarity , Struc- ture, Privacy Report Genera- tion Endoscopy Re- ports [105] 2024 Augmentation Neural Network Mistral, Llama, Gem...

2024
[242]

2024 Assistive W riting Neural Network T ransformers Publicly Avail- able (MIMIC- CXR) English ROUGE, BLEU, ME- TEOR, Human Assess- ment Similarity , Struc- ture Report Genera- tion Radiology

2024
[243]

2024 Assistive W riting Neural Network T ransformers (CLIP) Publicly Avail- able English BLEU, ROUGE, ME- TEOR, BER TScore Similarity , T est on Utility Report Gener- ation, Question Answering Radiology

2024
[244]

2024 Privacy- preserving Neural Net- work, Knowledge Source GPT (Mistral), RL, UML Publicly Avail- able (MIMIC-III) English SemScore Similarity , Pri- vacy - Discharge Sum- maries

2024
[245]

2024 Assistive W riting Neural Network T ransformers (PLM), GPT (Galactica) Publicly Avail- able English BLEU, ROUGE, ROUGE- L, METEOR Similarity Question Answer- ing Biological Con- cepts and Rela- tions

2024
[246]

2024 Usefulness Neural Network T ransformers (T5) Online source English Classification Accuracy , Human Assessment Structure, T est on Utility Named Entity Recognition Population, Intervention, Comparison, and Outcome (PICO)

2024
[247]

2024 Assistive W riting Neural Network LDM (V AE, CNN ResNet) Publicly Avail- able (IU X-RA Y, MIMIC-CXR) English BLEU, ROUGE-L Similarity , T est on Utility Report Genera- tion X-Ray , MRI, CT- scan

2024
[248]

2024 Usefulness Neural Network ChatGPT-3.5-turbo Prompting English Human Assessment Structure - History of Present Illness (HPI) Article Y ear Purpose Generation Method Approach / Architecture Data Source Language Evaluation Method Evaluation Paradigm Utility Clinical Source C...

2024
[249]

2024 Annotation, Cor- pus Building Neural Network GPT-4 Private EHR/EMR Norwegian Human Assessment, Clas- sification Accuracy Similarity , Struc- ture, T est on Utility De-Identification Discharge Sum- maries

2024
[250]

2024 Assistive W riting Neural Network T ransformers (VED) Publicly Avail- able (MIMIC- CXR) English BLEU, ROUGE-L, F1cXb Similarity , T est on Utility Report Genera- tion Radiology

2024
[251]

2024 Usefulness Neural Network ChatGPT-3.5 Publicly Avail- able (MIMIC- III), Prompting English Classification Accuracy T est on Utility Disease Classifi- cation Discharge Sum- maries

2024
[252]

2024 Usefulness Neural Network ChatGPT-3.5-T urbo, BioGPT, GPT-2, distil- GPT2, CerebroGPT Private EHR/EMR English BLEU, ROUGE, Cosine, TF-IDF, Human Assess- ment Similarity , Struc- ture, T est on Utility Report Genera- tion Radiology (Cere- brovascular)

2024
[253]

2024 Corpus Building Neural Network Claude-3-Opus, GPT-4 Publicly Avail- able English, Arabic ROUGE-L, BER TScore, Human Assessment Similarity , Struc- ture - Doctor-Patient Conversations

2024
[254]

2024 Corpus Building Neural Network GPT-4 Manual Collec- tion and Cura- tion, Prompting English ROUGE-L, Human Assess- ment, Classification Accu- racy Similarity , Struc- ture, T est on Utility Named Entity Recognition History of Present Illness (HPI)

2024
[255]

2024 Corpus Building Neural Network GPT (Llama) Publicly Avail- able English Human Assessment Structure, T est on Utility Question Answer- ing Discharge Sum- maries

2024
[256]

2025 Assistive W riting Neural Network T ransformers (BAR T) Publicly Avail- able English ROUGE, BLEU Similarity Question Answer- ing Doctor-Patient Conversations [106] 2025 Augmentation Neural Network Llama, Mistral, Gemma Publicly Avail- able English BLEU, ROUGE, ROUGE- L, W...

2025
[257]

2025 Augmentation, Annotation Neural Network ChatGPT-3.5 Private EHR/EMR, Prompting English Classification Accuracy , Pattern Match Structure, T est on Utility Phenotype Clas- sification History of Present Illness (HPI) Article Y ear Purpose Generation Method Approach / Archit...

2025
[2023]

Association for Computational Linguistics

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [55]

Scott H. Lee. Natural language generation for electronic health records. npj Digital Med , 1(1), nov 2018

2018

[2] [1]

Should free-text data in electronic medical records be shared for research? a citizens’ jury study in the UK

Elizabeth Ford, Malcolm Oswald, Lamiece Hassan, Kyle Bozentko, Goran Nenadic, and Jackie Cassell. Should free-text data in electronic medical records be shared for research? a citizens’ jury study in the UK. J Med Ethics , 46(6):367–377, may 2020

2020

[3] [2]

Synonym-based text generation in restructuring imbalanced dataset for deep learning models

Febi Siti Sutria Ningsih, Purnomo Husnul Khotimah, Andria Arisal, Andri Fachrur Rozie, Devi Munandar, Dianadewi Riswantini, Ekasari Nugraheni, Wiwin Suwarningsih, and Dian Kurniasari. Synonym-based text generation in restructuring imbalanced dataset for deep learning models. In 2022 5th International Conference on Networking, Information Systems and Secur...

2022

[4] [3]

Generating natural language adversarial examples on a large scale with generative models, 2020

Yankun Ren, Jianbin Lin, Siliang Tang, Jun Zhou, Shuang Yang, Yuan Qi, and Xiang Ren. Generating natural language adversarial examples on a large scale with generative models, 2020

2020

[5] [4]

Synthetic data generation: State of the art in health care domain

Hajra Murtaza, Musharif Ahmed, Naurin Farooq Khan, Ghulam Murtaza, Saad Zafar, and Ambreen Bano. Synthetic data generation: State of the art in health care domain. Computer Science Review, 48:100546, 2023

2023

[6] [5]

Synthetic data generation for tabular health records: A systematic review

Mikel Hernandez, Gorka Epelde, Ane Alberdi, Rodrigo Cilla, and Debbie Rankin. Synthetic data generation for tabular health records: A systematic review. Neurocomputing, 493:28–45, 2022

2022

[7] [6]

Deep generative models for synthetic data: A survey

Peter Eigenschink, Thomas Reutterer, Stefan Vamosi, Ralf Vamosi, Chang Sun, and Klaudius Kalcher. Deep generative models for synthetic data: A survey. IEEE Access, 11:47304–47320, 2023

2023

[8] [7]

A systematic literature review: deep learn- ing techniques for synthetic medical image generation and their applications in radiotherapy

Moiz Khan Sherwani and Shyam Gopalakrishnan. A systematic literature review: deep learn- ing techniques for synthetic medical image generation and their applications in radiotherapy. Frontiers in Radiology, 4:1385742, 2024

2024