Transform Blockchain into Distributed Parallel Computing Architecture for Precision Medicine

Abstract

This paper provides a vision and proposes mechanisms to transform the blockchain duplicated computing into distributed parallel computing architecture by transforming smart contract which features data driven from the ground up to support moving computing to native data strategy. This new distributed parallel computing architecture can be employed to build a large size of data set from various distributed hosted medical data sets which might consist of personal electronic medical record (EMR) and various medical data. This large medical data set will enable researchers to jump start the deep learning research for medical domain. Distributed data management, distributed data sharing, and distributed learning are the core mechanisms in the new architecture. The required new researches and developments to employ Google federated learning and transfer learning algorithms in this new architecture are discussed. The approach and mechanism enabled by the new architecture is illustrated to build a real world evidence of clinical trial toward personal and precision medicine. Research issues and technical challenges are provided.