REVIEW 7 cited by
Chest ImaGenome Dataset for Clinical Reasoning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Despite the progress in automatic detection of radiologic findings from chest X-ray (CXR) images in recent years, a quantitative evaluation of the explainability of these models is hampered by the lack of locally labeled datasets for different findings. With the exception of a few expert-labeled small-scale datasets for specific findings, such as pneumonia and pneumothorax, most of the CXR deep learning models to date are trained on global "weak" labels extracted from text reports, or trained via a joint image and unstructured text learning strategy. Inspired by the Visual Genome effort in the computer vision community, we constructed the first Chest ImaGenome dataset with a scene graph data structure to describe $242,072$ images. Local annotations are automatically produced using a joint rule-based natural language processing (NLP) and atlas-based bounding box detection pipeline. Through a radiologist constructed CXR ontology, the annotations for each CXR are connected as an anatomy-centered scene graph, useful for image-level reasoning and multimodal fusion applications. Overall, we provide: i) $1,256$ combinations of relation annotations between $29$ CXR anatomical locations (objects with bounding box coordinates) and their attributes, structured as a scene graph per image, ii) over $670,000$ localized comparison relations (for improved, worsened, or no change) between the anatomical locations across sequential exams, as well as ii) a manually annotated gold standard scene graph dataset from $500$ unique patients.
Forward citations
Cited by 7 Pith papers
-
AnatomiX, an Anatomy-Aware Grounded Multimodal Large Language Model for Chest X-Ray Interpretation
AnatomiX, a two-stage anatomy-first multimodal LLM for chest X-ray interpretation, reports >25% relative gains on anatomy grounding and grounded captioning, but some aggregate benchmark numbers are internally inconsis...
-
RadJEPA: Radiology Encoder for Chest X-Rays via Joint Embedding Predictive Architecture
RadJEPA learns chest X-ray encoders from unlabeled images via latent prediction in a joint embedding architecture, exceeding prior state-of-the-art on classification, segmentation, and report generation.
-
Knowledge to Sight: Reasoning over Visual Attributes via Knowledge Decomposition for Abnormality Grounding
Decomposing clinical terms into visual attributes lets 0.23B-2B vision-language models match or beat much larger medical VLMs for abnormality grounding with only 16k training pairs.
-
Interpreting Radiologist's Intention from Eye Movements in Chest X-ray Diagnosis
RadGazeIntent, a transformer model, predicts per-fixation diagnostic intention from radiologist gaze on chest X-rays, evaluated on three newly constructed intention-labeled datasets.
-
CheXPO: Preference Optimization for Chest X-ray VLMs with Counterfactual Rationale
A preference optimization strategy using confidence-based hard example mining, similarity retrieval, and synthetic counterfactual rationales improves chest X-ray VQA accuracy by 8.93% relative over supervised fine-tuning.
-
Learning to See Locally and Align Clinically with Pathology Semantics for Radiology Report Generation
Radiology report generation improves when image and text features are aligned through shared, CheXpert-initialized pathology prototypes and a masked-evidence objective; PALM reports state-of-the-art scores on three ch...
-
MCA-RG: Enhancing LLMs with Medical Concept Alignment for Radiology Report Generation
MCA-RG uses concept alignment, contrastive learning, matching loss, and feature gating to generate radiology reports, reporting SOTA on MIMIC-CXR and CheXpert Plus.
Discussion (0). Continue with ORCID to comment.