FedMood: Federated Learning on Mobile Health Data for Mood Detection

TL;DR

This paper introduces FedMood, a federated learning framework designed for privacy-preserving depression diagnosis using multi-source mobile health data, addressing data privacy issues in clinical AI applications.

Contribution

It proposes a general multi-view federated learning framework with late fusion methods for multi-source data, enabling privacy-preserving depression detection across institutions.

Findings

Federated learning achieves comparable accuracy to centralized models.

Late fusion effectively handles inconsistent multi-view time series.

Framework supports extension of traditional models to federated settings.

Abstract

Depression is one of the most common mental illness problems, and the symptoms shown by patients are not consistent, making it difficult to diagnose in the process of clinical practice and pathological research. Although researchers hope that artificial intelligence can contribute to the diagnosis and treatment of depression, the traditional centralized machine learning needs to aggregate patient data, and the data privacy of patients with mental illness needs to be strictly confidential, which hinders machine learning algorithms clinical application. To solve the problem of privacy of the medical history of patients with depression, we implement federated learning to analyze and diagnose depression. First, we propose a general multi-view federated learning framework using multi-source data, which can extend any traditional machine learning model to support federated learning across different institutions or parties. Secondly, we adopt late fusion methods to solve the problem of inconsistent time series of multi-view data. Finally, we compare the federated framework with other cooperative learning frameworks in performance and discuss the related results.