Researchers of Biomedical Electronics Center (BMEC) - Hanoi University of Science and Technology - have been creating a system to support the diagnosis of Hepatocellular Carcinoma (HCC) based on big data of ultrasound imaging and exome expression system. This is one of the main research directions on the application of artificial intelligence (AI) in healthcare that the center focuses on.

Speaking at AI4VN event organized at HUST from 15-16/8/2019, Assoc. Prof. Vu Duy Hai, Head of BMEC, said the study has applied number of new multi-tasking deep learning network architectures for detection and partition of HCC image,integrated expert information to navigate and bind target functions in the network, semantic information in training and network assessment. The researchers now can solve noise problems and mislabel images; therefore, they can build an intelligent software to support HCC diagnosis based on ultrasound images and tumor genome through DNA analysis from tissue samples to support accurate diagnosis. This will be the basis to determine the using the target drug for HCC patients. The diagnosis of HCC can be done from all medical facilities that have a diagnostic ultrasound machine and a computer or smartphone connected to the Internet. It is possible to improve the diagnosis and treatment of HCC, especially advanced HCC.

Moreover, with a large database built in the project, studies on auto-biology describing medical images and natural language-based searches in Vietnamese will also be implemented within the framework of the research.

HUST Communication and Branding Office