A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance

<b>Background:</b> Despite advances in artificial intelligence (AI) for Diabetic Retinopathy (DR) classification, traditional severity-based approaches often lack interpretability and fail to capture specific lesion-centered characteristics. To address these limitations, we constructed t...

Full description

Saved in:
Bibliographic Details
Main Authors: Gang-Min Park, Ji-Hoon Moon, Ho-Gil Jung
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Biomedicines
Subjects:
Online Access:https://www.mdpi.com/2227-9059/13/6/1446
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849433193490415616
author Gang-Min Park
Ji-Hoon Moon
Ho-Gil Jung
author_facet Gang-Min Park
Ji-Hoon Moon
Ho-Gil Jung
author_sort Gang-Min Park
collection DOAJ
description <b>Background:</b> Despite advances in artificial intelligence (AI) for Diabetic Retinopathy (DR) classification, traditional severity-based approaches often lack interpretability and fail to capture specific lesion-centered characteristics. To address these limitations, we constructed the National Medical Center (NMC) dataset, independently annotated by medical professionals with detailed labels of major DR lesions, including retinal hemorrhages, microaneurysms, and exudates. <b>Methods:</b> This study explores four critical research questions. First, we assess the analytical advantages of lesion-centered labeling compared to traditional severity-based labeling. Second, we investigate the potential complementarity between these labeling approaches through integration experiments. Third, we analyze how various model architectures and classification strategies perform under different labeling schemes. Finally, we evaluate decision-making differences between labeling methods using visualization techniques. We benchmarked the lesion-centered NMC dataset against the severity-based public Asia Pacific Tele-Ophthalmology Society (APTOS) dataset, conducting experiments with EfficientNet—a convolutional neural network architecture—and diverse classification strategies. <b>Results:</b> Our results demonstrate that binary classification effectively identifies severe non-proliferative Diabetic Retinopathy (Severe NPDR) exhibiting complex lesion patterns, while relationship-based learning enhances performance for underrepresented classes. Transfer learning from NMC to APTOS notably improved severity classification, achieving performance gains of 15.2% in mild cases and 66.3% in severe cases through feature fusion using Bidirectional Feature Pyramid Network (BiFPN) and Feature Pyramid Network (FPN). Visualization results confirmed that lesion-centered models focus more precisely on pathological features. <b>Conclusions:</b> Our findings highlight the benefits of integrating lesion-centered and severity-based information to enhance both accuracy and interpretability in DR classification. Future research directions include spatial lesion mapping and the development of clinically grounded learning methodologies.
format Article
id doaj-art-c183fc49109a4e8096161f428b99fd3f
institution Kabale University
issn 2227-9059
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Biomedicines
spelling doaj-art-c183fc49109a4e8096161f428b99fd3f2025-08-20T03:27:09ZengMDPI AGBiomedicines2227-90592025-06-01136144610.3390/biomedicines13061446A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and PerformanceGang-Min Park0Ji-Hoon Moon1Ho-Gil Jung2Department of Data Science, Seoul National University of Science and Technology, Seoul 01811, Republic of KoreaDepartment of Data Science, Seoul National University of Science and Technology, Seoul 01811, Republic of KoreaDepartment of Ophthalmology, National Medical Center, Seoul 04564, Republic of Korea<b>Background:</b> Despite advances in artificial intelligence (AI) for Diabetic Retinopathy (DR) classification, traditional severity-based approaches often lack interpretability and fail to capture specific lesion-centered characteristics. To address these limitations, we constructed the National Medical Center (NMC) dataset, independently annotated by medical professionals with detailed labels of major DR lesions, including retinal hemorrhages, microaneurysms, and exudates. <b>Methods:</b> This study explores four critical research questions. First, we assess the analytical advantages of lesion-centered labeling compared to traditional severity-based labeling. Second, we investigate the potential complementarity between these labeling approaches through integration experiments. Third, we analyze how various model architectures and classification strategies perform under different labeling schemes. Finally, we evaluate decision-making differences between labeling methods using visualization techniques. We benchmarked the lesion-centered NMC dataset against the severity-based public Asia Pacific Tele-Ophthalmology Society (APTOS) dataset, conducting experiments with EfficientNet—a convolutional neural network architecture—and diverse classification strategies. <b>Results:</b> Our results demonstrate that binary classification effectively identifies severe non-proliferative Diabetic Retinopathy (Severe NPDR) exhibiting complex lesion patterns, while relationship-based learning enhances performance for underrepresented classes. Transfer learning from NMC to APTOS notably improved severity classification, achieving performance gains of 15.2% in mild cases and 66.3% in severe cases through feature fusion using Bidirectional Feature Pyramid Network (BiFPN) and Feature Pyramid Network (FPN). Visualization results confirmed that lesion-centered models focus more precisely on pathological features. <b>Conclusions:</b> Our findings highlight the benefits of integrating lesion-centered and severity-based information to enhance both accuracy and interpretability in DR classification. Future research directions include spatial lesion mapping and the development of clinically grounded learning methodologies.https://www.mdpi.com/2227-9059/13/6/1446diabetic retinopathyfine-grained lesion detectionmedical expert labelingdataset integrationknowledge transferlesion-centered labeling
spellingShingle Gang-Min Park
Ji-Hoon Moon
Ho-Gil Jung
A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
Biomedicines
diabetic retinopathy
fine-grained lesion detection
medical expert labeling
dataset integration
knowledge transfer
lesion-centered labeling
title A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
title_full A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
title_fullStr A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
title_full_unstemmed A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
title_short A Comparative Study of Lesion-Centered and Severity-Based Approaches to Diabetic Retinopathy Classification: Improving Interpretability and Performance
title_sort comparative study of lesion centered and severity based approaches to diabetic retinopathy classification improving interpretability and performance
topic diabetic retinopathy
fine-grained lesion detection
medical expert labeling
dataset integration
knowledge transfer
lesion-centered labeling
url https://www.mdpi.com/2227-9059/13/6/1446
work_keys_str_mv AT gangminpark acomparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance
AT jihoonmoon acomparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance
AT hogiljung acomparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance
AT gangminpark comparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance
AT jihoonmoon comparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance
AT hogiljung comparativestudyoflesioncenteredandseveritybasedapproachestodiabeticretinopathyclassificationimprovinginterpretabilityandperformance