Future-Proofing Medical Imaging with Privacy-Preserving Federated Learning and Uncertainty Quantification: A Review

TL;DR

A comprehensive examination of federated learning, privacy-preserving FL (PPFL), and UQ in FL is provided, which identifies key gaps in current FL methodologies and proposes future research directions to enhance data privacy and trustworthiness in medical imaging applications.

Abstract

Artificial Intelligence (AI) has demonstrated significant potential in automating various medical imaging tasks, which could soon become routine in clinical practice for disease diagnosis, prognosis, treatment planning, and post-treatment surveillance. However, the privacy concerns surrounding patient data present a major barrier to the widespread adoption of AI in medical imaging, as large, diverse training datasets are essential for developing accurate, generalizable, and robust Artificial intelligence models. Federated Learning (FL) offers a solution that enables organizations to train AI models collaboratively without sharing sensitive data. federated learning exchanges model training information, such as gradients, between the participating sites. Despite its promise, federated learning is still in its developmental stages and faces several challenges. Notably, sensitive information can still be inferred from the gradients shared during model training. Quantifying AI models' uncertainty is vital due to potential data distribution shifts post-deployment, which can affect model performance. Uncertainty quantification (UQ) in FL is particularly challenging due to data heterogeneity across participating sites. This review provides a comprehensive examination of FL, privacy-preserving FL (PPFL), and UQ in FL. We identify key gaps in current FL methodologies and propose future research directions to enhance data privacy and trustworthiness in medical imaging applications.

Figures (4)

• Figure 1: An overview of federated learning (FL), PPFL, and UQ is presented. Combining FL with strong privacy-preservation and Uncertainty quantification methods can help the medical imaging community develop large-scale multi-institutional AI models that are truly generalizable, robust, and trustworthy.]
• Figure 2: An overview of FL algorithm types is presented. (A) In centralized federated learning, sites train a local model and pass the learned information to a central server to generate the global model, the global model is then passed to the local sites for further training. (B) Decentralized FL removes the need for a central server allowing for direct communication between sites. (C) Personalized FL leverages a central server while making a specific model for each site. Having a personalized model at each site is ideal in FL deployments with high data heterogeneity.
• Figure 3: A summary of privacy-preserving FL (PPFL) methods is presented. (A) Differential Privacy (DP) works by adding artificial noise into other gradient information before it is communicated, this hinders the ability of an attacker to extract useful information. (B) Homomorphic Encryption (HE) allows for mathematical operations to be performed on encrypted cyphertexts, and then once decrypted the results are as if the math was performed on plaintext. HE is useful in situations where the central server can't be trusted. (C) Various other methods of PPFL include hybrid approaches of DP and HE, knowledge transfer, loss differential strategies, and decentralized trust.
• Figure 4: A summary of UQ methods in FL is presented. (A) Model ensembling is where various models are trained and the final result is the average of their predictions. (B) Conformal Prediction (CP) is a method of UQ that provides a set of possible predictions, where the more uncertain the model is the more possible predictions it will provide. (C) Model calibration is a post-processing UQ method that serves to correct the issue of overconfidence in model prediction particularly when the model makes an incorrect prediction. This allows for more trustworthy confidence measures in the model's predictions. (D) Bayesian FL is another method of UQ that tracks the variance of the model during training and at inference time. The variance will go up as the model becomes more uncertain providing a measure of model uncertainty.