pith. sign in

arxiv: 2606.31502 · v1 · pith:QHFDSQJNnew · submitted 2026-06-30 · 💻 cs.CV

Fully Automated High-Precision Segmentation of Retinal Atrophy and Ellipsoid Zone Thickness in OCT: A Reliable Tool for Real-World GA Monitoring

Wolf-Dieter Vogl , Hlynur Skulason , Oliver Leingang , Ursula Schmidt-Erfurth , Amir Sadeghipour , Ariadne Whitby This is my paper

Pith reviewed 2026-07-01 06:04 UTC · model grok-4.3

classification 💻 cs.CV
keywords geographic atrophyOCT segmentationdeep learningretinal pigment epitheliumellipsoid zoneage-related macular degenerationbiomarker quantificationphotoreceptor degeneration
0
0 comments X

The pith

Three deep learning models segment RPE loss and EZ thickness in OCT volumes at sub-pixel precision for GA monitoring.

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

The paper presents a fully automated pipeline that deploys three specialized semantic segmentation models to delineate retinal pigment epithelium loss, ellipsoid zone boundaries including interruptions, and Bruch's membrane in spectral-domain OCT volumes. It reports Dice scores of 0.88 for RPE loss and 0.87 for EZ loss, Pearson correlation above 0.99, and average EZ thickness deviation of 2.15 micrometers, with reproducibility confirmed by ICC above 0.98 on a diverse training set of 298 volumes and an external validation set of 43 volumes spanning the AMD phenotypic spectrum. A sympathetic reader would care because consistent, pixel-wise quantification of outer photoreceptor degeneration and RPE loss would enable reliable tracking of geographic atrophy stage, progression rate, and treatment effects in both clinical trials and routine ophthalmic care.

Core claim

The proposed pipeline uses three specialized semantic segmentation models to delineate RPE loss, EZ boundaries (including interruptions), and Bruch's membrane. Results demonstrated high segmentation accuracy (Dice RPE loss: 0.88, Dice EZ loss: 0.87, Pearson's r > 0.99). Total EZ thickness measurements exhibited a sub-pixel average deviation of 2.15 μm, and segmentation reliability was confirmed by a strong reproducibility score (ICC > 0.98).

What carries the argument

Three specialized semantic segmentation models that separately delineate RPE loss, EZ boundaries including interruptions, and Bruch's membrane.

If this is right

  • High Dice scores and sub-pixel thickness accuracy enable detection of small changes in photoreceptor degeneration over short intervals.
  • ICC above 0.98 supports replacement of manual grading in clinical trials that require reproducible biomarker endpoints.
  • Robustness across lesion sizes and B-scan densities allows consistent measurements even when scan protocols vary.
  • Validation across GA, intermediate AMD, neovascular AMD, and healthy eyes supports use throughout the AMD disease spectrum.
  • Fully automated output removes inter-reader variability for routine monitoring of treatment response.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the models maintain accuracy on unseen scanner vendors, they could be embedded directly in commercial OCT consoles for immediate biomarker readouts.
  • The same segmentation approach might extend to tracking photoreceptor integrity in other macular diseases such as Stargardt or cone-rod dystrophies.
  • Longitudinal application could generate automated progression slopes that serve as surrogate endpoints for future GA trials.
  • Combining the outputs with existing layer segmentation tools could produce composite indices of outer retinal health without additional manual annotation.

Load-bearing premise

Performance on an independent external set of 43 volumes is sufficient to establish reliability across the full range of real-world GA cases and scanner variations.

What would settle it

A multi-center study on several hundred volumes acquired on different OCT devices showing Dice scores below 0.80 or EZ thickness deviations exceeding 5 μm in any major phenotypic subgroup.

Figures

Figures reproduced from arXiv: 2606.31502 by Amir Sadeghipour, Ariadne Whitby, Hlynur Skulason, Oliver Leingang, Ursula Schmidt-Erfurth, Wolf-Dieter Vogl.

Figure 1
Figure 1. Figure 1: Annotation example. En-face view (left) and central B-scan (right) showing retinal pigment epithelium (RPE) loss (blue) and ellipsoid zone (EZ) loss (green) annotations, as well as the layer annotations of inner boundary of the ellipsoid zone (IB-EZ) in light red, outer boundary of outer photoreceptors (OB-OPR) in dark red, and Bruch’s membrane (BM) in yellow. ellipsoid zone (IB-EZ), and outer boundary of … view at source ↗
Figure 2
Figure 2. Figure 2: Data flow diagram describing the inputs and outputs of each model. 2.3 Model training Three semantic segmentation models were developed to segment RPE loss, EZ and BM layer boundaries on SD-OCT scans ( [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of segmentation evaluation metrics comparing manual and automated segmentations of RPE loss (blue) and EZ loss (green) in the external validation dataset. Reported metrics are the absolute difference between measured and annotated lesion size, Dice similarity coefficient (DSC) for segmentation overlap, Hausdorff distance 95th percentile (HD95) 95% and average symmetric surface distance (ASSD) … view at source ↗
Figure 4
Figure 4. Figure 4: Bland Altman plot (top) and Deming regression fit (bottom) from automated segmen￾tation compared to manual ground truth annotation for RPE loss areas (left) and EZ loss areas (right). Numbers in brackets are the 95% confidence interval (CI). Inter-reader reliability [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Bland Altman plots for RPE loss area(left) and EZ loss area (right) inter-reader reliability comparing reader groups AN1 vs AN2 (top), AN1 vs automated segmentation (center) and AN2 vs automated segmentation (bottom). Numbers in brackets are the 95% confidence interval (CI) [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Reproducibility study: Bland Altman plot (top) and Deming regression fit (bottom) from automated segmentation of screening and baseline visit for RPE loss areas (left) and EZ loss areas (right). Numbers in brackets are the 95% confidence interval (CI) [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Effect of B-scan density on RPE loss and EZ loss measurement. Boxplot showing distribution of absolute errors (left) and absolute percentage errors (right) of RPE loss area, RPE loss progression area, EZ loss areas, and EZ loss progression area (top to bottom), comparing synthetically downsampled segmentations in the range of 19 to 128 B-Scans to ground-truth segmentation with 193 B-scans. 15 [PITH_FULL_I… view at source ↗
read the original abstract

Geographic atrophy (GA) secondary to age-related macular degeneration (AMD) requires precise monitoring of relevant structural biomarkers to assess disease stage, progression, and treatment response. This paper presents a fully automated, deep learning-based framework for the high-precision, pixel-wise segmentation of key biomarkers in optical coherence tomography (OCT) imaging: retinal pigment epithelium (RPE) loss, ellipsoid zone (EZ) loss, and EZ thinning. The proposed pipeline uses three specialized semantic segmentation models to delineate RPE loss, EZ boundaries (including interruptions), and Bruch's membrane. To ensure robustness and generalizability, the models were developed on a diverse dataset of 298 SD-OCT volumes representing the full phenotypic spectrum of AMD (GA:222, intermediate AMD: 40, neovascular AMD: 17, healthy: 19) and validated on an independent external dataset (n=43). The comprehensive evaluation was further strengthened using additional datasets to assess repeatability, inter-reader reliability, the impact of B-scan density on measurement accuracy, and subgroup performance stratified by lesion size. Results demonstrated high segmentation accuracy (Dice RPE loss: 0.88, Dice EZ loss: 0.87, Pearson's r > 0.99). Total EZ thickness measurements exhibited a sub-pixel average deviation of 2.15 $\mu m$, and segmentation reliability was confirmed by a strong reproducibility score (ICC > 0.98). By accurately and consistently quantifying outer photoreceptor degeneration and RPE loss, this fully automated framework provides a highly reliable tool for GA assessment in both clinical trials and routine real-world ophthalmic care.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript presents a fully automated deep learning pipeline employing three specialized semantic segmentation models to delineate RPE loss, EZ boundaries (including interruptions), and Bruch's membrane in SD-OCT volumes for geographic atrophy monitoring in AMD. The models are trained on a diverse set of 298 volumes (GA:222, iAMD:40, nAMD:17, healthy:19) and evaluated on an independent external set of 43 volumes, with additional tests for repeatability, inter-reader reliability, B-scan density effects, and subgroup performance by lesion size. Reported results include Dice scores of 0.88 (RPE loss) and 0.87 (EZ loss), Pearson's r > 0.99, average EZ thickness deviation of 2.15 μm, and ICC > 0.98, positioning the framework as a reliable tool for both clinical trials and real-world care.

Significance. If the generalizability claims hold under more rigorous external scrutiny, the work could deliver a standardized, high-precision automated tool for quantifying outer retinal degeneration biomarkers, reducing inter-observer variability and enabling consistent longitudinal monitoring of GA progression and treatment response.

major comments (1)
  1. [Abstract / External validation] Abstract and validation description: the external validation cohort (n=43) is substantially smaller than the training set (n=298) and the manuscript provides no explicit evidence that the external cohort matches the training distribution with respect to AMD subtype mix, lesion-size distribution, scanner vendors, or B-scan density. This gap directly undermines the load-bearing claim that the reported Dice scores, sub-pixel thickness error, and ICC > 0.98 establish the pipeline as a 'highly reliable tool for GA assessment in both clinical trials and routine real-world ophthalmic care' across the full phenotypic spectrum.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed review and constructive feedback. We address the major comment on external validation below, and will revise the manuscript accordingly to strengthen the presentation of generalizability.

read point-by-point responses

  1. Referee: [Abstract / External validation] Abstract and validation description: the external validation cohort (n=43) is substantially smaller than the training set (n=298) and the manuscript provides no explicit evidence that the external cohort matches the training distribution with respect to AMD subtype mix, lesion-size distribution, scanner vendors, or B-scan density. This gap directly undermines the load-bearing claim that the reported Dice scores, sub-pixel thickness error, and ICC > 0.98 establish the pipeline as a 'highly reliable tool for GA assessment in both clinical trials and routine real-world ophthalmic care' across the full phenotypic spectrum.

    Authors: We agree that the external cohort size (n=43) is smaller than the training set and that the manuscript does not currently include an explicit side-by-side comparison of distributions for AMD subtype, lesion size, scanner vendor, or B-scan density. The external set was acquired independently at a different clinical center to assess real-world applicability, but we acknowledge this does not substitute for transparent distribution matching. We will revise the manuscript to add a supplementary table (or expanded methods/results section) comparing these characteristics between training and external cohorts. We will also moderate the abstract and discussion claims to more precisely reflect the validated scope rather than implying coverage of the full phenotypic spectrum without qualification. revision: yes

Circularity Check

0 steps flagged

No significant circularity; validation is independent

full rationale

The paper trains three semantic segmentation models on a 298-volume dataset and reports direct performance metrics (Dice, Pearson r, ICC, thickness deviation) on a fully independent external 43-volume set plus repeatability cohorts. No equations, fitted parameters, or self-citations are invoked to derive the reported accuracies; the metrics are computed from model outputs versus ground truth without reduction to the training inputs by construction. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The claim depends on the empirical performance of trained neural networks on the described datasets, with no additional free parameters, axioms, or invented entities beyond standard assumptions in deep learning for image segmentation.

pith-pipeline@v0.9.1-grok · 5859 in / 1308 out tokens · 39749 ms · 2026-07-01T06:04:13.705112+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

64 extracted references · 58 canonical work pages · 1 internal anchor

  1. [1]

    Diagnostics , volume =

    Deep. Diagnostics , volume =. doi:10.3390/diagnostics15202580 , urldate =

    work page doi:10.3390/diagnostics15202580

  2. [2]

    Evaluating the

    Ansari, Georg and Sch. Evaluating the. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.66.11.34 , abstract =

    work page doi:10.1167/iovs.66.11.34

  3. [3]

    Exploring

    Birner, Klaudia and Mai, Julia and Boryshchuk, Daniela and Frommlet, Florian and Enzendorfer, Marie Louise and. Exploring. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.67.1.22 , urldate =

    work page doi:10.1167/iovs.67.1.22

  4. [4]

    Acta Ophthalmologica , volume =

    Normative Prospective Data on Automatically Quantified Retinal Morphology Correlated to Retinal Function in Healthy Ageing Eyes by Two Microperimetry Devices , author =. Acta Ophthalmologica , volume =. doi:10.1111/aos.17434 , urldate =

    work page doi:10.1111/aos.17434

  5. [5]

    and Steiner, Irene and Zarghami, Azin and Sadeghipour, Amir and

    Birner, Klaudia and Reiter, Gregor S. and Steiner, Irene and Zarghami, Azin and Sadeghipour, Amir and. Structure-. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.66.3.26 , urldate =

    work page doi:10.1167/iovs.66.3.26

  6. [6]

    Lancet (London, England) , volume =

    Statistical Methods for Assessing Agreement between Two Methods of Clinical Measurement , author =. Lancet (London, England) , volume =

  7. [7]

    Exact Parametric Confidence Intervals for

    Carkeet, Andrew , year = 2015, month = mar, journal =. Exact Parametric Confidence Intervals for. doi:10.1097/OPX.0000000000000513 , abstract =

    work page doi:10.1097/opx.0000000000000513 2015

  8. [8]

    and Johnston, Robert L

    Chakravarthy, Usha and Bailey, Clare C. and Johnston, Robert L. and McKibbin, Martin and Khan, Rehna S. and Mahmood, Sajjad and Downey, Louise and Dhingra, Narendra and Brand, Christopher and Brittain, Christopher J. and Willis, Jeffrey R. and Rabhi, Sarah and Muthutantri, Anushini and Cantrell, Ronald A. , year = 2018, month = jun, journal =. Characteriz...

    work page doi:10.1016/j.ophtha.2017.11.036 2018

  9. [9]

    and Reiter, Gregor S

    Coulibaly, Leonard M. and Reiter, Gregor S. and Fuchs, Philipp and Lachinov, Dmitrii and Leingang, Oliver and Vogl, Wolf-Dieter and Bogunovic, Hrvoje and. Progression. Ophthalmology Retina , volume =. doi:10.1016/j.oret.2023.05.004 , urldate =

    work page doi:10.1016/j.oret.2023.05.004 2023

  10. [10]

    Computer Methods and Programs in Biomedicine , volume =

    Fazekas, Botond and Aresta, Guilherme and Lachinov, Dmitrii and Riedl, Sophie and Mai, Julia and. Computer Methods and Programs in Biomedicine , volume =. doi:10.1016/j.cmpb.2025.108586 , urldate =

    work page doi:10.1016/j.cmpb.2025.108586 2025

  11. [11]

    Fazekas, Botond and Aresta, Guilherme and Lachinov, Dmitrii and Riedl, Sophie and Mai, Julia and. Medical. doi:10.1007/978-3-031-16452-1_31 , abstract =

    work page doi:10.1007/978-3-031-16452-1_31

  12. [12]

    Segmentation of

    Fazekas, Botond and Lachinov, Dmitrii and Aresta, Guilherme and Mai, Julia and. Segmentation of. IEEE Journal of Biomedical and Health Informatics , volume =. doi:10.1109/JBHI.2022.3217962 , urldate =

    work page doi:10.1109/jbhi.2022.3217962 2022

  13. [13]

    and Yehoshua, Zohar and Gregori, Giovanni and Penha, Fernando M

    Feuer, William J. and Yehoshua, Zohar and Gregori, Giovanni and Penha, Fernando M. and Chew, Emily Y. and Ferris, Frederick L. and Clemons, Traci E. and Lindblad, Anne S. and Rosenfeld, Philip J. , year = 2013, month = jan, journal =. Square. doi:10.1001/jamaophthalmol.2013.572 , urldate =

    work page doi:10.1001/jamaophthalmol.2013.572 2013

  14. [14]

    Investigative Ophthalmology & Visual Science , volume =

    Quantifications of. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.66.4.65 , urldate =

    work page doi:10.1167/iovs.66.4.65

  15. [15]

    Gerendas, Bianca S. and Sadeghipour, Amir and Michl, Martin and Goldbach, Felix and Mylonas, Georgios and Gruber, Anastasiia and Alten, Thomas and Leingang, Oliver and Sacu, Stefan and Bogunovic, Hrvoje and. Validation of an. RETINA , volume =. doi:10.1097/IAE.0000000000003557 , urldate =

    work page doi:10.1097/iae.0000000000003557

  16. [16]

    and Lad, Eleonora M

    Heier, Jeffrey S. and Lad, Eleonora M. and Holz, Frank G. and Rosenfeld, Philip J. and Guymer, Robyn H. and Boyer, David and Grossi, Federico and Baumal, Caroline R. and Korobelnik, Jean-Francois and Slakter, Jason S. and Waheed, Nadia K. and Metlapally, Ravi and Pearce, Ian and Steinle, Nathan and Francone, Anibal A. and Hu, Allen and Lally, David R. and...

    work page doi:10.1016/s0140-6736(23)01520-9

  17. [17]

    and Strauss, Erich C

    Holz, Frank G. and Strauss, Erich C. and. Geographic. Ophthalmology , volume =. doi:10.1016/j.ophtha.2013.11.023 , urldate =

    work page doi:10.1016/j.ophtha.2013.11.023 2013

  18. [18]

    Densely connected convolutional networks

    Huang, Gao and Liu, Zhuang and Van Der Maaten, Laurens and Weinberger, Kilian Q. , year = 2017, month = jul, pages =. Densely. 2017. doi:10.1109/CVPR.2017.243 , urldate =

    work page doi:10.1109/cvpr.2017.243 2017

  19. [19]

    and Srivastava, Sunil K

    Kalra, Gagan and Cetin, Hasan and Whitney, Jon and Yordi, Sari and Cakir, Yavuz and McConville, Conor and Whitmore, Victoria and Bonnay, Michelle and Reese, Jamie L. and Srivastava, Sunil K. and Ehlers, Justis P. , year = 2023, month = mar, journal =. Automated. doi:10.3390/diagnostics13061178 , urldate =

    work page doi:10.3390/diagnostics13061178 2023

  20. [20]

    and Poon, Wesley and Bawiskar, Anaya and Nehra, Natasha and Davarci, Orhun and Das, Glori and Vasquez, Matthew and Schaal, Shlomit and Raghunathan, Raksha and Wong, Stephen T

    Kermany, Daniel S. and Poon, Wesley and Bawiskar, Anaya and Nehra, Natasha and Davarci, Orhun and Das, Glori and Vasquez, Matthew and Schaal, Shlomit and Raghunathan, Raksha and Wong, Stephen T. C. , year = 2025, month = jun, journal =. Identifying. doi:10.1167/iovs.66.6.55 , urldate =

    work page doi:10.1167/iovs.66.6.55 2025

  21. [21]

    Koo and Mae Y

    Koo, Terry K. and Li, Mae Y. , year = 2016, month = jun, journal =. A. doi:10.1016/j.jcm.2016.02.012 , urldate =

    work page doi:10.1016/j.jcm.2016.02.012 2016

  22. [22]

    and Frank, Sophie and Coulibaly, Leonard M

    Kostolna, Klaudia and Reiter, Gregor S. and Frank, Sophie and Coulibaly, Leonard M. and Fuchs, Philipp and R. A. Ophthalmology Science , volume =. doi:10.1016/j.xops.2023.100456 , urldate =

    work page doi:10.1016/j.xops.2023.100456 2023

  23. [23]

    Projective

    Lachinov, Dmitrii and Seeb. Projective. Medical. doi:10.1007/978-3-030-87193-2_41 , urldate =

    work page doi:10.1007/978-3-030-87193-2_41

  24. [24]

    and Heitkotter, Heather and Carroll, Joseph , year = 2021, month = oct, journal =

    Lee, Karen E. and Heitkotter, Heather and Carroll, Joseph , year = 2021, month = oct, journal =. Challenges. doi:10.1167/tvst.10.12.27 , urldate =

    work page doi:10.1167/tvst.10.12.27 2021

  25. [25]

    Graefe's Archive for Clinical and Experimental Ophthalmology , volume =

    Sequential Structural and Functional Change in Geographic Atrophy on Multimodal Imaging in Non-Exudative Age-Related Macular Degeneration , author =. Graefe's Archive for Clinical and Experimental Ophthalmology , volume =. doi:10.1007/s00417-023-06022-3 , urldate =

    work page doi:10.1007/s00417-023-06022-3

  26. [26]

    and Grossi, Federico V

    Liao, David S. and Grossi, Federico V. and El Mehdi, Delphine and Gerber, Monica R. and Brown, David M. and Heier, Jeffrey S. and Wykoff, Charles C. and Singerman, Lawrence J. and Abraham, Prema and Grassmann, Felix and Nuernberg, Peter and Weber, Bernhard H. F. and Deschatelets, Pascal and Kim, Robert Y. and Chung, Carol Y. and Ribeiro, Ramiro M. and Ham...

    work page doi:10.1016/j.ophtha.2019.07.011 2020

  27. [27]

    Liefers, Bart and Colijn, Johanna M. and. A. Ophthalmology , volume =. doi:10.1016/j.ophtha.2020.02.009 , abstract =

    work page doi:10.1016/j.ophtha.2020.02.009 2020

  28. [28]

    Clinical Chemistry , volume =

    Evaluation of Regression Procedures for Methods Comparison Studies , author =. Clinical Chemistry , volume =. doi:10.1093/clinchem/39.3.424 , urldate =

    work page doi:10.1093/clinchem/39.3.424

  29. [29]

    Dual-Branch Image Projection Network for Geographic Atrophy Segmentation in Retinal

    Liu, Xiaoming and Li, Jieyang and Zhang, Ying and Yao, Junping , year = 2025, month = feb, journal =. Dual-Branch Image Projection Network for Geographic Atrophy Segmentation in Retinal. doi:10.1038/s41598-025-90709-6 , urldate =

    work page doi:10.1038/s41598-025-90709-6 2025

  30. [30]

    and Vogl, Wolf-Dieter and Bogunovic, Hrvoje and

    Mai, Julia and Lachinov, Dmitrii and Riedl, Sophie and Reiter, Gregor S. and Vogl, Wolf-Dieter and Bogunovic, Hrvoje and. Clinical Validation for Automated Geographic Atrophy Monitoring on. Scientific Reports , volume =. doi:10.1038/s41598-023-34139-2 , urldate =

    work page doi:10.1038/s41598-023-34139-2

  31. [31]

    and Riedl, Sophie and Vogl, Wolf-Dieter and Sadeghipour, Amir and McKeown, Alex and Foos, Emma and Bogunovic, Hrvoje and

    Mai, Julia and Reiter, Gregor S. and Riedl, Sophie and Vogl, Wolf-Dieter and Sadeghipour, Amir and McKeown, Alex and Foos, Emma and Bogunovic, Hrvoje and. Dynamics of the. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.66.11.48 , urldate =

    work page doi:10.1167/iovs.66.11.48

  32. [32]

    and Riedl, Sophie and Vogl, Wolf-Dieter and Sadeghipour, Amir and Foos, Emma and McKeown, Alex and Bogunovic, Hrvoje and

    Mai, Julia and Reiter, Gregor S. and Riedl, Sophie and Vogl, Wolf-Dieter and Sadeghipour, Amir and Foos, Emma and McKeown, Alex and Bogunovic, Hrvoje and. Quantitative Comparison of Automated. Scientific Reports , volume =. doi:10.1038/s41598-024-71496-y , urldate =

    work page doi:10.1038/s41598-024-71496-y

  33. [33]

    and Gumpinger, Markus and Leigang, Oliver and Bogunovic, Hrvoje and Barthelmes, Daniel and Nehemy, Marcio B

    Mares, Virginia and Reiter, Gregor S. and Gumpinger, Markus and Leigang, Oliver and Bogunovic, Hrvoje and Barthelmes, Daniel and Nehemy, Marcio B. and. Correlation of Retinal Fluid and Photoreceptor and. Acta Ophthalmologica , volume =. doi:10.1111/aos.16799 , urldate =

    work page doi:10.1111/aos.16799

  34. [34]

    , year = 2025, month = jun, journal =

    Midroni, Julie and Longwell, Jack and Bhambra, Nishaant and Demian, Sueellen and Pecaku, Aurora and Melo, Isabela Martins and Muni, Rajeev H. , year = 2025, month = jun, journal =. Automated Segmentation of Subretinal Fluid from Optical Coherence Tomography:. doi:10.1016/j.xops.2025.100852 , urldate =

    work page doi:10.1016/j.xops.2025.100852 2025

  35. [35]

    and Hu, Zhihong , year = 2020, month = jun, journal =

    Mishra, Zubin and Ganegoda, Anushika and Selicha, Jane and Wang, Ziyuan and Sadda, SriniVas R. and Hu, Zhihong , year = 2020, month = jun, journal =. Automated. doi:10.1038/s41598-020-66355-5 , urldate =

    work page doi:10.1038/s41598-020-66355-5 2020

  36. [36]

    Mon. The. Translational Vision Science & Technology , volume =. doi:10.1167/tvst.7.6.40 , urldate =

    work page doi:10.1167/tvst.7.6.40

  37. [37]

    Morano, Jos. Deep. IEEE Journal of Biomedical and Health Informatics , volume =. doi:10.1109/JBHI.2024.3352970 , urldate =

    work page doi:10.1109/jbhi.2024.3352970 2024

  38. [38]

    Self-Supervised

    Morano, Jos. Self-Supervised. Medical. doi:10.1007/978-3-031-43901-8_56 , abstract =

    work page doi:10.1007/978-3-031-43901-8_56

  39. [39]

    Automated

    Orlando, Jos. Automated. Scientific Reports , volume =. doi:10.1038/s41598-020-62329-9 , urldate =

    work page doi:10.1038/s41598-020-62329-9

  40. [40]

    Orlando, Jos. U2-. doi:10.48550/ARXIV.1901.07929 , urldate =

    work page doi:10.48550/arxiv.1901.07929 1901

  41. [41]

    Association of Complement

    Pfau, Maximilian and. Association of Complement. Scientific Reports , volume =. doi:10.1038/s41598-022-22404-9 , urldate =

    work page doi:10.1038/s41598-022-22404-9

  42. [42]

    Progression of

    Pfau, Maximilian and. Progression of. JAMA Ophthalmology , volume =. doi:10.1001/jamaophthalmol.2020.2914 , urldate =

    work page doi:10.1001/jamaophthalmol.2020.2914 2020

  43. [43]

    Pramil, Varsha and. A. Ophthalmology Retina , volume =. doi:10.1016/j.oret.2022.08.007 , urldate =

    work page doi:10.1016/j.oret.2022.08.007 2022

  44. [44]

    and Mai, Julia and Riedl, Sophie and Birner, Klaudia and Frank, Sophie and Bogunovic, Hrvoje and

    Reiter, Gregor S. and Mai, Julia and Riedl, Sophie and Birner, Klaudia and Frank, Sophie and Bogunovic, Hrvoje and. Progress in Retinal and Eye Research , volume =. doi:10.1016/j.preteyeres.2024.101305 , abstract =

    work page doi:10.1016/j.preteyeres.2024.101305 2024

  45. [45]

    and Told, Reinhard and Schranz, Markus and Baumann, Lukas and Mylonas, Georgios and Sacu, Stefan and Pollreisz, Andreas and

    Reiter, Gregor S. and Told, Reinhard and Schranz, Markus and Baumann, Lukas and Mylonas, Georgios and Sacu, Stefan and Pollreisz, Andreas and. Subretinal. Investigative Ophthalmology & Visual Science , volume =. doi:10.1167/iovs.61.6.11 , urldate =

    work page doi:10.1167/iovs.61.6.11

  46. [46]

    and Lachinov, Dmitrii and Grechenig, Christoph and McKeown, Alex and Scheibler, Lukas and Bogunovi

    Riedl, Sophie and Vogl, Wolf-Dieter and Mai, Julia and Reiter, Gregor S. and Lachinov, Dmitrii and Grechenig, Christoph and McKeown, Alex and Scheibler, Lukas and Bogunovi. The. Ophthalmology Retina , volume =. doi:10.1016/j.oret.2022.05.030 , urldate =

    work page doi:10.1016/j.oret.2022.05.030 2022

  47. [47]

    Journal of Ophthalmology , volume =

    Automatic. Journal of Ophthalmology , volume =. doi:10.1155/2020/8204641 , urldate =

    work page doi:10.1155/2020/8204641 2020

  48. [48]

    Wightman, Ross , year = 2019, publisher =

    2019

  49. [49]

    and Guymer, Robyn and Holz, Frank G

    Sadda, Srinivas R. and Guymer, Robyn and Holz, Frank G. and. Consensus. Ophthalmology , volume =. doi:10.1016/j.ophtha.2017.09.028 , urldate =

    work page doi:10.1016/j.ophtha.2017.09.028 2017

  50. [50]

    and Rabe, Christina and Schiffman, Courtney and Yang, Qi and Lee, Aaron Y

    Salvi, Anish and Cluceru, Julia and Gao, Simon S. and Rabe, Christina and Schiffman, Courtney and Yang, Qi and Lee, Aaron Y. and Keane, Pearse A. and Sadda, Srinivas R. and Holz, Frank G. and Ferrara, Daniela and Anegondi, Neha , year = 2025, month = mar, journal =. Deep. doi:10.1016/j.xops.2024.100635 , urldate =

    work page doi:10.1016/j.xops.2024.100635 2025

  51. [51]

    Ophthalmology , volume =

    Disease. Ophthalmology , volume =. doi:10.1016/j.ophtha.2024.08.017 , urldate =

    work page doi:10.1016/j.ophtha.2024.08.017 2024

  52. [52]

    Investigative Ophthalmology & Visual Science , volume =

    Longitudinal Assessment of Progressive Functional Decline in Areas of Intact Retina,. Investigative Ophthalmology & Visual Science , volume =

  53. [53]

    , year = 2024, journal =

    Seghier, Mohamed L. , year = 2024, journal =. Image. doi:10.1002/ima.23203 , urldate =

    work page doi:10.1002/ima.23203 2024

  54. [54]

    and Gim, Nayoon and Blazes, Marian and Lee, Cecilia S

    Spaide, Theodore and Rajesh, Anand E. and Gim, Nayoon and Blazes, Marian and Lee, Cecilia S. and Macivannan, Niranchana and Lee, Gary and Lewis, Warren and Salehi, Ali and de Sisternes, Luis and Herrera, Gissel and Shen, Mengxi and Gregori, Giovanni and Rosenfeld, Philip J. and Pramil, Varsha and Waheed, Nadia and Wu, Yue and Zhang, Qinqin and Lee, Aaron ...

    work page doi:10.1016/j.xops.2024.100587 2025

  55. [55]

    Navigating Distribution Shifts in Medical Image Analysis: A Survey

    Su, Zixian and Guo, Jingwei and Yang, Xi and Wang, Qiufeng and Coenen, Frans and Huang, Kaizhu , year = 2025, month = aug, number =. Navigating. doi:10.48550/arXiv.2411.05824 , urldate =. arXiv , keywords =:2411.05824 , primaryclass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2411.05824 2025

  56. [56]

    Proceedings of the 36th

    Tan, Mingxing and Le, Quoc , year = 2019, month = may, pages =. Proceedings of the 36th

    2019

  57. [57]

    Investigative Ophthalmology & Visual Science , volume =

    Automated. Investigative Ophthalmology & Visual Science , volume =

  58. [58]

    and Lachinov, Dmitrii and Bogunovi

    Vogl, Wolf-Dieter and Riedl, Sophie and Mai, Julia and Reiter, Gregor S. and Lachinov, Dmitrii and Bogunovi. Predicting. Ophthalmology Retina , volume =. doi:10.1016/j.oret.2022.08.003 , urldate =

    work page doi:10.1016/j.oret.2022.08.003 2022

  59. [59]

    Spatio-Temporal Alterations in Retinal and Choroidal Layers in the Progression of Age-Related Macular Degeneration (

    Vogl, Wolf-Dieter and Bogunovi. Spatio-Temporal Alterations in Retinal and Choroidal Layers in the Progression of Age-Related Macular Degeneration (. Scientific Reports , volume =. doi:10.1038/s41598-021-85110-y , urldate =

    work page doi:10.1038/s41598-021-85110-y

  60. [60]

    Wightman, Ross , year = 2025, month = jul, doi =

    2025

  61. [61]

    Journal of Medical Imaging , volume =

    Family of Boundary Overlap Metrics for the Evaluation of Medical Image Segmentation , author =. Journal of Medical Imaging , volume =. doi:10.1117/1.JMI.5.1.015006 , urldate =

    work page doi:10.1117/1.jmi.5.1.015006

  62. [62]

    and Cukras, Catherine and Rabe, Christina and Ferrara, Daniela and Spaide, Richard F

    Yoshida, Kenta and Anegondi, Neha and Pely, Adam and Zhang, Miao and Debraine, Frederic and Ramesh, Karthik and Steffen, Verena and Gao, Simon S. and Cukras, Catherine and Rabe, Christina and Ferrara, Daniela and Spaide, Richard F. and Sadda, SriniVas R. and Holz, Frank G. and Yang, Qi , year = 2025, month = feb, journal =. Deep. doi:10.1167/tvst.14.2.11 ...

    work page doi:10.1167/tvst.14.2.11 2025

  63. [63]

    Photoreceptor

    Zekavat, Seyedeh Maryam and Sekimitsu, Sayuri and Ye, Yixuan and Raghu, Vineet and Zhao, Hongyu and Elze, Tobias and Segr. Photoreceptor. Ophthalmology , volume =. doi:10.1016/j.ophtha.2022.02.001 , abstract =

    work page doi:10.1016/j.ophtha.2022.02.001 2022

  64. [64]

    doi:10.1109/TMI.2019.2959609 , urldate =

    Zhou, Zongwei and Siddiquee, Md Mahfuzur Rahman and Tajbakhsh, Nima and Liang, Jianming , year = 2020, month = jun, journal =. doi:10.1109/TMI.2019.2959609 , urldate =

    work page doi:10.1109/tmi.2019.2959609 2020