The Extensible Data-Brain Model: Architecture, Applications and Directions

Abstract One of the key ideas in realizing human-like intelligence is to understand enough information-processing mechanisms in the human brain. Brain Informatics is a rapidly expanding interdisciplinary field to systematically utilize brain-related data, information, and knowledge coming from the entire research process for deeply brain investigation. In the past few years, a data-centric conceptual brain model, namely Data-Brain, has been proposed to meet requirements of a systematic methodology of Brain Informatics. Although the Data-Brain model provides a conceptual framework and detailed description for managing and analyzing brain big data, the increasing demand still requires the support of existing and future advanced technologies. The goal of this paper is to discuss and explore the extensible version of the Data-Brain with advanced computing techniques, which will benefit high-efficiency brain data exploitation and facilitate high-efficient brain data management. Particularly, we advocate that the human-in-the-loop approach should be designed in the processes of data mining and knowledge discovery using human-computer interaction and collaborative computing with human intelligence. Their synergistic use is expected to power future progress for building intelligent systems and applications connected with the study of complex human brain.