Deep Learning to Predict the Future Growth of Geographic Atrophy from Fundus Autofluorescence

IF 3.2 Q1 OPHTHALMOLOGY

Ophthalmology science Pub Date : 2024-10-23 DOI:10.1016/j.xops.2024.100635

Anish Salvi MS , Julia Cluceru PhD , Simon S. Gao PhD , Christina Rabe PhD , Courtney Schiffman PhD , Qi Yang PhD , Aaron Y. Lee MD, MSCI , Pearse A. Keane MD, FRCOphth , Srinivas R. Sadda MD , Frank G. Holz MD , Daniela Ferrara MD, PhD , Neha Anegondi MTech

{"title":"Deep Learning to Predict the Future Growth of Geographic Atrophy from Fundus Autofluorescence","authors":"Anish Salvi MS , Julia Cluceru PhD , Simon S. Gao PhD , Christina Rabe PhD , Courtney Schiffman PhD , Qi Yang PhD , Aaron Y. Lee MD, MSCI , Pearse A. Keane MD, FRCOphth , Srinivas R. Sadda MD , Frank G. Holz MD , Daniela Ferrara MD, PhD , Neha Anegondi MTech","doi":"10.1016/j.xops.2024.100635","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><div>The region of growth (ROG) of geographic atrophy (GA) throughout the macular area has an impact on visual outcomes. Here, we developed multiple deep learning models to predict the 1-year ROG of GA lesions using fundus autofluorescence (FAF) images.</div></div><div><h3>Design</h3><div>In this retrospective analysis, 3 types of models were developed using FAF images collected 6 months after baseline to predict the GA lesion area (segmented lesion mask) at 1.5 years, FAF images collected at baseline and 6 months to predict the GA lesion at 1.5 years, and FAF images collected 6 months after baseline to predict the GA lesion at 1 and 1.5 years. The 1-year ROG from the 6-month visit was derived by taking the difference between the GA lesion area (segmented lesion mask) at the 1.5-year and 6-month visits.</div></div><div><h3>Participants</h3><div>Patients enrolled in the following lampalizumab clinical trials and prospective observational studies: NCT02247479, NCT02247531, NCT02479386, and NCT02399072.</div></div><div><h3>Methods</h3><div>Datasets of study eyes from 597 patients were split into model training (310), validation (78), and test sets (209), stratified by baseline or initial lesion area, lesion growth rate, foveal involvement, and focality. Deep learning experiments were performed using the 2-dimensional U-Net; whole-lesion and multiclass models were developed.</div></div><div><h3>Main Outcome Measures</h3><div>The performance of the models was evaluated by calculating the Dice score, coefficient of determination (R<sup>2</sup>), and the squared Pearson correlation coefficient (r<sup>2</sup>) between the true and derived GA lesion 1-year ROG.</div></div><div><h3>Results</h3><div>The model using baseline and 6-month FAF images to predict GA lesion enlargement at 1.5 years had the best performance for the derived 1-year ROG. Mean Dice scores were 0.73, 0.68, and 0.70 in the training, validation, and test sets, respectively. The R<sup>2</sup> (0.77, 0.53, and 0.79) and r<sup>2</sup> (0.83, 0.61, and 0.79) showed similar trends across the 3 sets.</div></div><div><h3>Conclusions</h3><div>These findings show the potential of using baseline and/or 6-month visit FAF images to predict 1-year GA ROG using a deep learning approach. This work could potentially help support decision-making in clinical trials and more informed treatment decisions in clinical practice.</div></div><div><h3>Financial Disclosure(s)</h3><div>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 2","pages":"Article 100635"},"PeriodicalIF":3.2000,"publicationDate":"2024-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11699103/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ophthalmology science","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666914524001714","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose

The region of growth (ROG) of geographic atrophy (GA) throughout the macular area has an impact on visual outcomes. Here, we developed multiple deep learning models to predict the 1-year ROG of GA lesions using fundus autofluorescence (FAF) images.

Design

In this retrospective analysis, 3 types of models were developed using FAF images collected 6 months after baseline to predict the GA lesion area (segmented lesion mask) at 1.5 years, FAF images collected at baseline and 6 months to predict the GA lesion at 1.5 years, and FAF images collected 6 months after baseline to predict the GA lesion at 1 and 1.5 years. The 1-year ROG from the 6-month visit was derived by taking the difference between the GA lesion area (segmented lesion mask) at the 1.5-year and 6-month visits.

Participants

Patients enrolled in the following lampalizumab clinical trials and prospective observational studies: NCT02247479, NCT02247531, NCT02479386, and NCT02399072.

Methods

Datasets of study eyes from 597 patients were split into model training (310), validation (78), and test sets (209), stratified by baseline or initial lesion area, lesion growth rate, foveal involvement, and focality. Deep learning experiments were performed using the 2-dimensional U-Net; whole-lesion and multiclass models were developed.

Main Outcome Measures

The performance of the models was evaluated by calculating the Dice score, coefficient of determination (R²), and the squared Pearson correlation coefficient (r²) between the true and derived GA lesion 1-year ROG.

Results

The model using baseline and 6-month FAF images to predict GA lesion enlargement at 1.5 years had the best performance for the derived 1-year ROG. Mean Dice scores were 0.73, 0.68, and 0.70 in the training, validation, and test sets, respectively. The R² (0.77, 0.53, and 0.79) and r² (0.83, 0.61, and 0.79) showed similar trends across the 3 sets.

Conclusions

These findings show the potential of using baseline and/or 6-month visit FAF images to predict 1-year GA ROG using a deep learning approach. This work could potentially help support decision-making in clinical trials and more informed treatment decisions in clinical practice.

Financial Disclosure(s)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

查看原文本刊更多论文

基于眼底自体荧光的深度学习预测地理萎缩的未来增长。

目的：地理萎缩（GA）的生长区域（ROG）遍及黄斑区域，影响视力结果。在这里，我们开发了多个深度学习模型，利用眼底自身荧光（FAF）图像预测GA病变1年的ROG。设计：在本回顾性分析中，采用基线后6个月收集的FAF图像来预测1.5年的GA病变面积（分段病变遮罩），基线后6个月收集的FAF图像来预测1.5年的GA病变，基线后6个月收集的FAF图像来预测1年和1.5年的GA病变，建立了3种模型。6个月随访后的1年ROG是通过1.5年和6个月随访时GA病变面积（分段病变掩膜）的差异得出的。参与者：纳入以下lampalizumab临床试验和前瞻性观察性研究的患者：NCT02247479， NCT02247531， NCT02479386和NCT02399072。方法：来自597例患者的研究眼数据集被分为模型训练集（310）、验证集（78）和测试集（209），并按基线或初始病变面积、病变生长速度、中央凹受累程度和焦点进行分层。采用二维U-Net进行深度学习实验；建立了全病变模型和多级模型。主要结局指标：通过计算Dice评分、决定系数（R2）以及真实和衍生GA病变1年ROG之间的Pearson相关系数（R2）的平方来评估模型的性能。结果：使用基线和6个月FAF图像预测1.5年时GA病变扩大的模型对衍生的1年ROG具有最佳性能。在训练集、验证集和测试集中，平均Dice得分分别为0.73、0.68和0.70。3组间的R2（0.77、0.53、0.79）和R2（0.83、0.61、0.79）的变化趋势相似。结论：这些发现表明使用基线和/或6个月就诊FAF图像使用深度学习方法预测1年GA ROG的潜力。这项工作可能有助于支持临床试验中的决策，并在临床实践中做出更明智的治疗决策。财务披露：专有或商业披露可在本文末尾的脚注和披露中找到。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊