The bottleneck of applying artificial intelligence (AI) algorithm to optimize physical devices is to acquire a large amount of data under realistic conditions. The data is essential for any modeling and is known as the “ground truth”. However, it is extremely challenging to find or acquire biomedicalrelevant data regarding the biomechanics and physiology of a human body. Compared with data in other fields, there is a sparsity of data and no efficient means to obtain ground truths due to safety, ethical, legal and technical reasons. To solve this problem, in this project, we are going to build the cyber-physical twin of the human organ, which contains means to acquire the necessary data, and to represent this data as a hybrid (physical and cyber) model that mimics application-oriented aspects of the human physiology, including organs. The physical part represents the human anatomy and the complex properties of the real biological material of an organ, and more importantly it also includes an embedded novel soft sensor network and can relay information and data beyond what can be obtained from a real human organ. The abundant data generated from the physical part will be analyzed in a computational system. Facilitated with existing physical and medical knowledge, a cyber model of the organ will then be built. The digital model will allow data-driven simulations with altered experimental conditions to minimize the time and effort of hardware-based experiments on physical models. The first aim is to build a human brain model. The cyber-physical brain will not only be useful for medical applications, such as the testing of surgical instruments and the training of medical personnel; it will also serve as a unique development tool for human-machine hardware interfaces, such as helmets and automotive safety systems. The project is funded by the Cyber Valley Research Fund.
|Project Number||PN 2A-2|
|Alternative Project Number||CyVy-RF-2020-09|
|Project Name||The Cyber-Physical Twin of the Human Organ|
|Project Duration||May 2020 - June 2023|
|Project Leader||Dr. Tian Qiu|
|Project Members||Moonkwang Jeong, PhD Researcher
Do Yeon Kim, PhD Researcher