Ayesha Jabbar, Jianjun Huang, Muhammad Kashif Jabbar, Asad Ali
{"title":"Adaptive Multimodal Fusion in Vertical Federated Learning for Decentralized Glaucoma Screening.","authors":"Ayesha Jabbar, Jianjun Huang, Muhammad Kashif Jabbar, Asad Ali","doi":"10.3390/brainsci15090990","DOIUrl":null,"url":null,"abstract":"<p><p><b>Background/Objectives:</b> Early and accurate detection of glaucoma is vital for preventing irreversible vision loss, yet traditional diagnostic approaches relying solely on unimodal retinal imaging are limited by data sparsity and constrained context. Furthermore, real-world clinical data are often fragmented across institutions under strict privacy regulations, posing significant challenges for centralized machine learning methods. <b>Methods:</b> To address these barriers, this study proposes a novel Quality Aware Vertical Federated Learning (QAVFL) framework for decentralized multimodal glaucoma detection. The proposed system dynamically integrates clinical text, retinal fundus images, and biomedical signal data through modality-specific encoders, followed by a Fusion Attention Module (FAM) that adaptively weighs the reliability and contribution of each modality. Unlike conventional early fusion or horizontal federated learning methods, QAVFL operates in vertically partitioned environments and employs secure aggregation mechanisms incorporating homomorphic encryption and differential privacy to preserve patient confidentiality. <b>Results:</b> Extensive experiments conducted under heterogeneous non-IID settings demonstrate that QAVFL achieves an accuracy of 98.6%, a recall of 98.6%, an F1-score of 97.0%, and an AUC of 0.992, outperforming unimodal and early fusion baselines with statistically significant improvements (<i>p</i> < 0.01). <b>Conclusions:</b> The findings validate the effectiveness of dynamic multimodal fusion under privacy-preserving decentralized learning and highlight the scalability and clinical applicability of QAVFL for robust glaucoma screening across fragmented healthcare environments.</p>","PeriodicalId":9095,"journal":{"name":"Brain Sciences","volume":"15 9","pages":""},"PeriodicalIF":2.8000,"publicationDate":"2025-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12468838/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Brain Sciences","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.3390/brainsci15090990","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"NEUROSCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
Background/Objectives: Early and accurate detection of glaucoma is vital for preventing irreversible vision loss, yet traditional diagnostic approaches relying solely on unimodal retinal imaging are limited by data sparsity and constrained context. Furthermore, real-world clinical data are often fragmented across institutions under strict privacy regulations, posing significant challenges for centralized machine learning methods. Methods: To address these barriers, this study proposes a novel Quality Aware Vertical Federated Learning (QAVFL) framework for decentralized multimodal glaucoma detection. The proposed system dynamically integrates clinical text, retinal fundus images, and biomedical signal data through modality-specific encoders, followed by a Fusion Attention Module (FAM) that adaptively weighs the reliability and contribution of each modality. Unlike conventional early fusion or horizontal federated learning methods, QAVFL operates in vertically partitioned environments and employs secure aggregation mechanisms incorporating homomorphic encryption and differential privacy to preserve patient confidentiality. Results: Extensive experiments conducted under heterogeneous non-IID settings demonstrate that QAVFL achieves an accuracy of 98.6%, a recall of 98.6%, an F1-score of 97.0%, and an AUC of 0.992, outperforming unimodal and early fusion baselines with statistically significant improvements (p < 0.01). Conclusions: The findings validate the effectiveness of dynamic multimodal fusion under privacy-preserving decentralized learning and highlight the scalability and clinical applicability of QAVFL for robust glaucoma screening across fragmented healthcare environments.
期刊介绍:
Brain Sciences (ISSN 2076-3425) is a peer-reviewed scientific journal that publishes original articles, critical reviews, research notes and short communications in the areas of cognitive neuroscience, developmental neuroscience, molecular and cellular neuroscience, neural engineering, neuroimaging, neurolinguistics, neuropathy, systems neuroscience, and theoretical and computational neuroscience. Our aim is to encourage scientists to publish their experimental and theoretical results in as much detail as possible. There is no restriction on the length of the papers. The full experimental details must be provided so that the results can be reproduced. Electronic files or software regarding the full details of the calculation and experimental procedure, if unable to be published in a normal way, can be deposited as supplementary material.