Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease
Abstract Stargardt disease type 1 (STGD1) is a genetic disorder that leads to progressive vision loss, with no approved treatments currently available. The development of effective therapies faces the challenge of identifying appropriate outcome measures that accurately reflect treatment benefits. O...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-85213-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823862339940646912 |
|---|---|
| author | Parisa Khateri Tiana Koottungal Damon Wong Rupert W. Strauss Lucas Janeschitz-Kriegl Maximilian Pfau Leopold Schmetterer Hendrik P. N. Scholl |
| author_facet | Parisa Khateri Tiana Koottungal Damon Wong Rupert W. Strauss Lucas Janeschitz-Kriegl Maximilian Pfau Leopold Schmetterer Hendrik P. N. Scholl |
| author_sort | Parisa Khateri |
| collection | DOAJ |
| description | Abstract Stargardt disease type 1 (STGD1) is a genetic disorder that leads to progressive vision loss, with no approved treatments currently available. The development of effective therapies faces the challenge of identifying appropriate outcome measures that accurately reflect treatment benefits. Optical Coherence Tomography (OCT) provides high-resolution retinal images, serving as a valuable tool for deriving potential outcome measures, such as retinal thickness. However, automated segmentation of OCT images, particularly in regions disrupted by degeneration, remains complex. In this study, we propose a deep learning-based approach that incorporates a pathology-aware loss function to segment retinal sublayers in OCT images from patients with STGD1. This method targets relatively unaffected regions for sublayer segmentation, ensuring accurate boundary delineation in areas with minimal disruption. In severely affected regions, identified by a box detection model, the total retina is segmented as a single layer to avoid errors. Our model significantly outperforms standard models, achieving an average Dice coefficient of $$99\%$$ for total retina and $$93\%$$ for retinal sublayers. The most substantial improvement was in the segmentation of the photoreceptor inner segment, with Dice coefficient increasing by $$25\%$$ . This approach provides a balance between granularity and reliability, making it suitable for clinical application in tracking disease progression and evaluating therapeutic efficacy. |
| format | Article |
| id | doaj-art-cb1f7dd91106466581a7e4e8470d87f5 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-cb1f7dd91106466581a7e4e8470d87f52025-02-09T12:34:22ZengNature PortfolioScientific Reports2045-23222025-02-0115111410.1038/s41598-025-85213-wLooking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt diseaseParisa Khateri0Tiana Koottungal1Damon Wong2Rupert W. Strauss3Lucas Janeschitz-Kriegl4Maximilian Pfau5Leopold Schmetterer6Hendrik P. N. Scholl7Institute of Molecular and Clinical Ophthalmology BaselInstitute of Molecular and Clinical Ophthalmology BaselInstitute of Molecular and Clinical Ophthalmology BaselInstitute of Molecular and Clinical Ophthalmology BaselInstitute of Molecular and Clinical Ophthalmology BaselDepartment of Ophthalmology, University of BaselInstitute of Molecular and Clinical Ophthalmology BaselDepartment of Clinical Pharmacology, Medical University of ViennaAbstract Stargardt disease type 1 (STGD1) is a genetic disorder that leads to progressive vision loss, with no approved treatments currently available. The development of effective therapies faces the challenge of identifying appropriate outcome measures that accurately reflect treatment benefits. Optical Coherence Tomography (OCT) provides high-resolution retinal images, serving as a valuable tool for deriving potential outcome measures, such as retinal thickness. However, automated segmentation of OCT images, particularly in regions disrupted by degeneration, remains complex. In this study, we propose a deep learning-based approach that incorporates a pathology-aware loss function to segment retinal sublayers in OCT images from patients with STGD1. This method targets relatively unaffected regions for sublayer segmentation, ensuring accurate boundary delineation in areas with minimal disruption. In severely affected regions, identified by a box detection model, the total retina is segmented as a single layer to avoid errors. Our model significantly outperforms standard models, achieving an average Dice coefficient of $$99\%$$ for total retina and $$93\%$$ for retinal sublayers. The most substantial improvement was in the segmentation of the photoreceptor inner segment, with Dice coefficient increasing by $$25\%$$ . This approach provides a balance between granularity and reliability, making it suitable for clinical application in tracking disease progression and evaluating therapeutic efficacy.https://doi.org/10.1038/s41598-025-85213-wStargardt DiseaseOptical Coherence TomographyDeep LearningRetina SegmentationPathology-Aware Loss FunctionAutomated Image Analysis |
| spellingShingle | Parisa Khateri Tiana Koottungal Damon Wong Rupert W. Strauss Lucas Janeschitz-Kriegl Maximilian Pfau Leopold Schmetterer Hendrik P. N. Scholl Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease Scientific Reports Stargardt Disease Optical Coherence Tomography Deep Learning Retina Segmentation Pathology-Aware Loss Function Automated Image Analysis |
| title | Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease |
| title_full | Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease |
| title_fullStr | Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease |
| title_full_unstemmed | Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease |
| title_short | Looking outside the box with a pathology aware AI approach for analyzing OCT retinal images in Stargardt disease |
| title_sort | looking outside the box with a pathology aware ai approach for analyzing oct retinal images in stargardt disease |
| topic | Stargardt Disease Optical Coherence Tomography Deep Learning Retina Segmentation Pathology-Aware Loss Function Automated Image Analysis |
| url | https://doi.org/10.1038/s41598-025-85213-w |
| work_keys_str_mv | AT parisakhateri lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT tianakoottungal lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT damonwong lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT rupertwstrauss lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT lucasjaneschitzkriegl lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT maximilianpfau lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT leopoldschmetterer lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease AT hendrikpnscholl lookingoutsidetheboxwithapathologyawareaiapproachforanalyzingoctretinalimagesinstargardtdisease |