The paper focuses on the role of federated learning in a healthcare environment. The experimental setup involved different healthcare providers, each with their datasets. A comparison was made between training a deep learning model using traditional methods, where all the data is stored in one place, and using federated learning, where the data is distributed among the workers. The experiment aimed to identify possible challenges that could arise when training a model in a federated learning scenario, including the impact of federated learning on the obtained measures for breast density classification and examining the impact of data preprocessing and domain adaptation. The results indicate that using federated learning deep-learning models can be effectively trained on distributed healthcare data, performing similarly to the traditional approach while providing additional benefits such as improved data privacy and security. However, domain adaptation and data heterogeneity must be carefully addressed to achieve optimal performance in federated learning.
Authors
Additional information
- DOI
- Digital Object Identifier link open in new tab 10.1109/icit58465.2023.10143132
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2023
