Privacy-Preserving Federated Depression Detection From Multisource Mobile Health Data
| Autor: | Lifang He, Lianzhong Liu, Zhifeng Hao, Zakirul Alam Bhuiyan, Hao Peng, Xiaohang Xu, Lichao Sun |
|---|---|
| Rok vydání: | 2022 |
| Předmět: |
Cooperative learning
Information privacy Computer science business.industry Process (engineering) Mental illness medicine.disease Machine learning computer.software_genre Computer Science Applications Control and Systems Engineering medicine Confidentiality Medical history Artificial intelligence Electrical and Electronic Engineering business computer Multi-source Depression (differential diagnoses) Information Systems |
| Zdroj: | IEEE Transactions on Industrial Informatics. 18:4788-4797 |
| ISSN: | 1941-0050 1551-3203 |
| Popis: | 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. The experimental results show that in the case of participating in learning, when there are enough participants, the prediction accuracy of depression score can reach 85.13%, which is about 15% higher than that of local training. When the number of participants is small and the amount of data is sufficient, the prediction accuracy rate of depression score can also reach 84.32%, and the improvement rate is about 9%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|