Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FedMood: Federated Learning on Mobile Health Data for Mood Detection

Xiaohang Xu, Hao Peng, Lichao Sun, Md Zakirul Alam Bhuiyan, Lianzhong Liu, Lifang He

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.

FedMood: Federated Learning on Mobile Health Data for Mood Detection

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.

Paper Structure

This paper contains 20 sections, 10 equations, 3 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background
  3. Related Work
  4. Later Fusion Model
  5. Iot-data silo island problem and methodology
  6. Problem Description
  7. Federated learning
  8. Experiments
  9. Dataset
  10. Experimental Setup
  11. IID Experiments
  12. Compared Methods
  13. Evaluation criteria
  14. Experiment Result
  15. Non-IID Experiments
  16. ...and 5 more sections

Figures (3)

  • Figure 1: A comparison of different strategies for fusing multi-view data from the perspective of computational graph cao2017deepmood.
  • Figure 2: The architecture of federated learning. Firstly, the participated parties have data on normal people, bipolar I and bipolar II users, and no data interaction between different parties. At the beginning of each communication round, the server will assign the global model to the participated party in training in this round. Next, the activated parties will train the local model through its own mobile health data and upload it to the server. Finally, the server updates the global model according to the uploaded local model.
  • Figure 3: Visualization of labeling with TSNE for three views