Dataset Information

A foundation model for generalizable disease detection from retinal images.

ABSTRACT: Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders¹. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications². Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from unlabelled retinal images and provides a basis for label-efficient model adaptation in several applications. Specifically, RETFound is trained on 1.6 million unlabelled retinal images by means of self-supervised learning and then adapted to disease detection tasks with explicit labels. We show that adapted RETFound consistently outperforms several comparison models in the diagnosis and prognosis of sight-threatening eye diseases, as well as incident prediction of complex systemic disorders such as heart failure and myocardial infarction with fewer labelled data. RETFound provides a generalizable solution to improve model performance and alleviate the annotation workload of experts to enable broad clinical AI applications from retinal imaging.

SUBMITTER: Zhou Y

PROVIDER: S-EPMC10550819 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A foundation model for generalizable disease detection from retinal images.

Zhou Yukun Y Chia Mark A MA Wagner Siegfried K SK Ayhan Murat S MS Williamson Dominic J DJ Struyven Robbert R RR Liu Timing T Xu Moucheng M Lozano Mateo G MG Woodward-Court Peter P Kihara Yuka Y Altmann Andre A Lee Aaron Y AY Topol Eric J EJ Denniston Alastair K AK Alexander Daniel C DC Keane Pearse A PA

Nature 20230913 7981

Medical artificial intelligence (AI) offers great potential for recognizing signs of health conditions in retinal images and expediting the diagnosis of eye diseases and systemic disorders<sup>1</sup>. However, the development of AI models requires substantial annotation and models are usually task-specific with limited generalizability to different clinical applications<sup>2</sup>. Here, we present RETFound, a foundation model for retinal images that learns generalizable representations from u ...[more]

PMID: 37704728

Similar Datasets

Project description:BackgroundThe 12-lead electrocardiogram (ECG) remains a cornerstone of cardiac diagnostics, yet existing artificial intelligence (AI) solutions for automated interpretation often lack generalizability, remain closed-source, and are primarily trained using supervised learning, limiting their adaptability across diverse clinical settings. To address these challenges, we developed and compared two open-source foundational ECG models: DeepECG-SSL, a self-supervised learning model, and DeepECG-SL, a supervised learning model.MethodsBoth models were trained on over 1 million ECGs using a standardized preprocessing pipeline and automated free-text extraction from ECG reports to predict 77 cardiac conditions. DeepECG-SSL was pretrained using self-supervised contrastive learning and masked lead modeling. The models were evaluated on six multilingual private healthcare systems and four public datasets for ECG interpretation across 77 diagnostic categories. Fairness analyses assessed disparities in performance across age and sex groups, while also investigating fairness and resource utilization.ResultsDeepECG-SSL achieved AUROCs of 0.990 (95%CI 0.990, 0.990) on internal dataset, 0.981 (95%CI 0.981, 0.981) on external public datasets, and 0.983 (95%CI 0.983, 0.983) on external private datasets, while DeepECG-SL demonstrated AUROCs of 0.992 (95%CI 0.992, 0.992), 0.980 (95%CI 0.980, 0.980) and 0.983 (95%CI 0.983, 0.983) respectively. Fairness analyses revealed minimal disparities (true positive rate & false positive rate difference<0.010) across age and sex groups. Digital biomarker prediction (Long QT syndrome (LQTS) classification, 5-year atrial fibrillation prediction and left ventricular ejection fraction (LVEF) classification) with limited labeled data, DeepECG-SSL outperformed DeepECG-SL in predicting 5-year atrial fibrillation risk (N=132,050; AUROC 0.742 vs. 0.720; Δ=0.022; P<0.001), identifying reduced LVEF ≤40% (N=25,252; 0.928 vs. 0.900; Δ=0.028; P<0.001), and classifying LQTS syndrome subtypes (N=127; 0.931 vs. 0.853; Δ=0.078; P=0.026).ConclusionBy releasing model weights, preprocessing tools, and validation code, we aim to support robust, data-efficient AI diagnostics across diverse clinical environments. This study establishes self-supervised learning as a promising paradigm for ECG analysis, particularly in settings with limited annotated data, enhancing accessibility, generalizability, and fairness in AI-driven cardiac diagnostics.

Project description:ObjectivesThe retinal age gap (RAG) is emerging as a potential biomarker for various diseases of the human body, yet its utility depends on machine learning models capable of accurately predicting biological retinal age from fundus images. However, training generalizable models is hindered by potential shortages of diverse training data. To overcome these obstacles, this work develops a novel and computationally efficient distributed learning framework for retinal age prediction.Materials and methodsThe proposed framework employs a memory-efficient 8-bit quantized version of RETFound, a cutting-edge foundation model for retinal image analysis, to extract features from fundus images. These features are then used to train an efficient linear regression head model for predicting retinal age. The framework explores federated learning (FL) as well as traveling model (TM) approaches for distributed training of the linear regression head. To evaluate this framework, we simulate a client network using fundus image data from the UK Biobank. Additionally, data from patients with type 1 diabetes from the UK Biobank and the Brazilian Multilabel Ophthalmological Dataset (BRSET) were utilized to explore the clinical utility of the developed methods.ResultsOur findings reveal that the developed distributed learning framework achieves retinal age prediction performance on par with centralized methods, with FL and TM providing similar performance (mean absolute error of 3.57 ± 0.18 years for centralized learning, 3.60 ± 0.16 years for TM, and 3.63 ± 0.19 years for FL). Notably, the TM was found to converge with fewer local updates than FL. Moreover, patients with type 1 diabetes exhibited significantly higher RAG values than healthy controls in all models, for both the UK Biobank and BRSET datasets (P < .001).DiscussionThe high computational and memory efficiency of the developed distributed learning framework makes it well suited for resource-constrained environments.ConclusionThe capacity of this framework to integrate data from underrepresented populations for training of retinal age prediction models could significantly enhance the accessibility of the RAG as an important disease biomarker.

Project description:Nuclear DNA is organized into a compact three-dimensional (3D) structure that impacts critical cellular processes. High-throughput chromosome conformation capture (Hi-C) is the most widely used method for measuring 3D genome architecture, while linear epigenomic assays, such as ATAC-seq, DNase-seq, and ChIP-seq, are extensively employed to characterize epigenomic regulation. However, the integrative analysis of chromatin interactions and associated epigenomic regulation remains challenging due to the pairwise nature of Hi-C data, mismatched resolution between Hi-C and epigenomic assays, and inconsistencies among analysis tools. Here we propose HiCFoundation, a Hi-C-based foundation model for integrative analysis linking chromatin structure to downstream regulatory function. HiCFoundation is trained from hundreds of Hi-C assays encompassing 118 million contact matrix submatrices. The model achieves state-of-the-art performance on multiple types of 3D genome analysis, including reproducibility analysis, resolution enhancement, and loop detection. We further demonstrate the model's generalizability through genome architecture analysis of 316 species. Notably, by enhancing low-coverage experimental Hi-C data, HiCFoundation reveals genome-wide loop loss during differentiation of hematopoietic stem and progenitor cells (HSPCs) to neutrophils. Additionally, HiCFoundation is able to predict multiple types of epigenomic activity from Hi-C input and further interprets the link between Hi-C input and epigenomic output to reveal the relationship between chromatin conformation and genome function. Finally, HiCFoundation can analyze single-cell Hi-C data, shedding light on genome structure at single-cell resolution. HiCFoundation thus provides a unified, efficient, generalizable, and interpretable foundation for genome architecture, single-cell and multi-omics analysis across species, paving the path for systematically studying genome 3D architecture and its regulatory mechanisms.

Dataset Information

A foundation model for generalizable disease detection from retinal images.

Publications

A foundation model for generalizable disease detection from retinal images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets