| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 3.8 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Current data protection regulations restrict cross-institutional data sharing impeding collaborative efforts in predictive disease diagnostics. We explore blockchain-based distributed machine learning, a promising approach to privacy-preserving collaborative learning between health institutions to advance disease diagnostics and research. Expanding on earlier work, we provide an overview of previously introduced models and develop SGD Chain, which implements a Stochastic Gradient Descent classifier and proposes a novel approach to sharing model parameters combining Smart Contracts and IPFS. The results prove the technical feasibility of SGD Chain. Based thereon, we discuss the potential of monetizing the output model to provide financial incentives for participants.
Descrição
Palavras-chave
Machine learning Business analytics Healthcare Blockchain Predictive disease diagnostics
