Swarm Learning to identify COVID-19, tuberculosis and leukemia patients based on blood transcriptomes

Dac ID Contact Person Email Access Information
EGAC00001001639 Thomas Ulas t [dot] ulas [at] uni-bonn [dot] de No additional information is available

This DAC controls 1 dataset:

Dataset ID Description Technology Samples
EGAD00001006231 Identification of patients with life-threatening diseases including leukemias or infections such as tuberculosis or COVID-19 is an important goal of modern precision medicine. However, there is an increasing divide between what is technically possible and what is allowed because of privacy legislation. We have recently illustrated that classical machine learning can identify leukemia patients based on their blood transcriptomes. To facilitate integration of any omics data from any data owner world-wide without violating privacy laws, we here introduce Swarm Learning (SL), a decentralized machine learning approach uniting edge computing, artificial intelligence (AI), blockchain and privacy protection without the need for a central coordinator thereby going beyond federated learning. To illustrate its feasibility, using more than 12,000 transcriptomes from peripheral blood mononuclear cells and more than 2,000 peripheral blood transcriptomes we demonstrate that SL of omics data distributed across different individual sites leads to disease classifiers that outperform those developed at individual sites. Yet, SL completely protects local privacy regulations by design. We propose this approach to noticeably accelerate the introduction of precision medicine. NextSeq 500 650