Federated attention consistent learning models for prostate cancer diagnosis and Gleason grading

TL;DR

This paper introduces a federated learning framework with attention consistency and differential privacy for prostate cancer diagnosis and grading, achieving high accuracy across multiple centers without sharing raw data.

Contribution

The study presents a novel federated attention-consistent learning (FACL) framework that improves model generalization and privacy preservation in multi-center prostate cancer pathology analysis.

Findings

FACL achieved an AUC of 0.9718 in diagnosis, outperforming individual centers.

FACL attained a Kappa score of 0.8463 in Gleason grading, surpassing other models.

The framework effectively balances accuracy and data privacy in medical AI applications.

Abstract

Artificial intelligence (AI) holds significant promise in transforming medical imaging, enhancing diagnostics, and refining treatment strategies. However, the reliance on extensive multicenter datasets for training AI models poses challenges due to privacy concerns. Federated learning provides a solution by facilitating collaborative model training across multiple centers without sharing raw data. This study introduces a federated attention-consistent learning (FACL) framework to address challenges associated with large-scale pathological images and data heterogeneity. FACL enhances model generalization by maximizing attention consistency between local clients and the server model. To ensure privacy and validate robustness, we incorporated differential privacy by introducing noise during parameter transfer. We assessed the effectiveness of FACL in cancer diagnosis and Gleason grading tasks using 19,461 whole-slide images of prostate cancer from multiple centers. In the diagnosis task, FACL achieved an area under the curve (AUC) of 0.9718, outperforming seven centers with an average AUC of 0.9499 when categories are relatively balanced. For the Gleason grading task, FACL attained a Kappa score of 0.8463, surpassing the average Kappa score of 0.7379 from six centers. In conclusion, FACL offers a robust, accurate, and cost-effective AI training model for prostate cancer pathology while maintaining effective data safeguards.