Federated learning for multi-center imaging diagnostics: a simulation study in cardiovascular disease

Abstract

Deep learning models can enable accurate and efficient disease diagnosis, but have thus far been hampered by the data scarcity present in the medical world. Automated diagnosis studies have been constrained by underpowered single-center datasets, and although some results have shown promise, their generalizability to other institutions remains questionable as the data heterogeneity between institutions is not taken into account. By allowing models to be trained in a distributed manner that preserves patients’ privacy, federated learning promises to alleviate these issues, by enabling diligent multi-center studies. We present the first simulated federated learning study on the modality of cardiovascular magnetic resonance and use four centers derived from subsets of the M&M and ACDC datasets, focusing on the diagnosis of hypertrophic cardiomyopathy. We adapt a 3D-CNN network pretrained on action recognition and explore two different ways of incorporating shape prior information to the model, and four different data augmentation set-ups, systematically analyzing their impact on the different collaborative learning choices. We show that despite the small size of data (180 subjects derived from four centers), the privacy preserving federated learning achieves promising results that are competitive with traditional centralized learning. We further find that federatively trained models exhibit increased robustness and are more sensitive to domain shift effects.

Federated learning enables big data for rare cancer boundary detection

“…This study is meant to be used as an example for future FL studies between collaborators with an inherent amount of trust that can result in clinically deployable ML models. Further research is required to assess privacy concerns in a detailed manner 63,64 and to apply FL to different tasks and data types [66][67][68][69] . Building on this study, a continuous FL consortium would enable downstream quantitative analyses with implications for both routine practice and clinical trials, and most importantly, increase access to high-quality precision care worldwide.…”

“…68 Department of Neurosurgery, Case Western Reserve University School of Medicine, Cleveland, OH, USA. 69 National Cancer Institute, National Institute of Health, Division of Cancer Epidemiology and Genetics, Bethesda, MD, USA. 70 Department of Radiology, Neuroradiology Division, University of Pittsburgh, Pittsburgh, PA, USA.…”

Open problems in medical federated learning

“…Using patients' computer tomography (CT) and magnetic resonance imaging images as data, machine learning models predict fatal diseases connected to the patient's life. Linardos, for example, preprocessed M&M and ACDC Dataset as N4 via field correction showed higher accuracy than in the DML environment (Linardos et al, 2022). They used federated learning with the ResNet model for hypertrophic cardiomyopathy diagnosis.…”

“…Linardos, for example, preprocessed M&M and ACDC Dataset as N4 via field correction showed higher accuracy than in the DML environment (Linardos et al , 2022). They used federated learning with the ResNet model for hypertrophic cardiomyopathy diagnosis.…”

Where do we stand in AI for endoscopic image analysis? Deciphering gaps and future directions

“…It is, however, difficult to establish what number of test samples provide non-skewed results. While unseen test sets determine the generalisability of approaches, most supervised techniques designed are not robust to unseen data distributions 106 . Hence, generalisability assessments or robustness tests are often not included in most papers.…”