JMIR Publications

ABSTRACT

Background:

Machine learning (ML) and artificial intelligence (AI) have shown promising results in many areas and are driven by the increasing amount of available data. However, this data is often distributed across different institutions and cannot be easily shared due to strict privacy regulations. Federated learning (FL) allows for the training of distributed ML models without sharing sensitive data. Also, the implementation is time-consuming and requires advanced programming skills and complex technical infrastructures.

Objective:

Various tools and frameworks have been published to simplify the development of FL algorithms and provide the necessary technical infrastructure. While there are many high-quality frameworks, most of them focus on only a single application case or method. To our knowledge, there are no generic frameworks, meaning that existing solutions are restricted to a particular kind of algorithm or application field. Furthermore, most of these frameworks provide an API that needs programming knowledge. There is no collection of ready-to-use FL algorithms that is extendable and allows users (e.g., researchers) without programming knowledge to apply FL. A central FL platform for both FL algorithm developers and users does not exist.

Methods:

To address this gap and make FL available to everyone, we developed FeatureCloud, an all-in-one platform for FL in biomedicine and beyond. The platform consists of three main components: a global frontend, a global backend, and a local controller. Our platform uses docker to separate local acting components of the platform from sensitive data systems. We evaluate our platform with two different algorithms on four datasets for both accuracy and runtime.

Results:

FeatureCloud removes the complexity of distributed systems for developers and end-users by providing a comprehensive platform for executing multi-institutional FL analyses and implementing FL algorithms. Through its integrated AI Store, federated algorithms can be easily published and re-used by the community. To secure sensitive raw data, FeatureCloud supports privacy-enhancing technologies to secure the shared local models and assures high standards in data privacy to comply with the strict GDPR. Our evaluation shows that apps developed in FeatureCloud can produce highly similar results compared to centralized approaches and scale well for an increasing number of participating sites.

Conclusions:

FeatureCloud is the first ready-to-use platform that integrates the development and execution of FL algorithms while reducing the complexity to a minimum and removing the hurdles of federated infrastructure. We thus believe that it has the potential to vastly increase the accessibility of privacy-preserving and distributed data analysis in biomedicine and beyond.