In order to promote the cross research of information, intelligence and life and health and serve the major national strategy of "facing people's life and health", the "Cross Series Academic Salon of Information, Intelligence and Life and Health" (phase 1), jointly sponsored by CAAI and CAA, was successfully held online on January 14, 2022. The theme of this salon is "Biomedical Big Data and Artificial Intelligence: Now and Future". It is jointly organized by CAAI bioinformatics and artificial life special committee, CAA Special Committee on intelligent health and biological information, Department of automation of Tsinghua University, and co-organized by Beijing National Research Center of information science and technology, School of Aeronautics and Astronautics of Xiamen University, Fuzhou Institute of data technology and other units.

Zhang Xuegong, chairman of CAAI bioinformatics and artificial life special committee (hereinafter referred to as "the committee") and professor of Tsinghua University, Li Xia, Professor of Harbin Medical University and Hainan Medical University, Wang Jianxin, Professor of Central South University and Dean of computer college, Gao Lin, Professor of Xidian University, Guo Maozu, Professor of Beijing University of Civil Engineering and Architecture and Dean of School of electrical and information engineering, Professor Huang Deshuang of Tongji University and Professor Shen Hongbin of Shanghai Jiaotong University were invited to participate in the activity. Professor Wang Ying of Xiamen University, Chen Gang, founder of wegene, and Peng Tao, chief data scientist of yiduyun attended the meeting as seminar guests. The keynote speech session was hosted by Professor Wang Xiaowo of Tsinghua University, and the seminar session was hosted by Gu Jin, Secretary General of the committee and associate professor of Tsinghua University.

In the keynote speech session, Zhang Xuegong gave a report entitled "A review of research on pattern recognition of biological information". He reviewed his experience of cutting into the field of bioinformatics from the problem of machine learning with small samples more than 20 years ago, and shared his preliminary exploration on this issue. Li Xia made a thematic report entitled "biomedical big data: 'big data' + 'big health', pointing out that it is very important to put forward needs from medicine and solve practical problems. Due to the high complexity of the human body, traditional machine learning methods have many new challenges in the face of biomedical problems.

 Wang Jianxin made a thematic report entitled "thinking on the application of biomedical big data", in which puts forward six challenges in biomedical big data research, namely data integration and integration, data governance, open data sharing, transparency and reproducibility, interpretability, and data-driven learning of integrated knowledge. Huang Deshuang made a thematic report entitled "thinking about the development of biomedical big data and artificial intelligence" points out that neural networks have great potential in the field of biomedical big data. Traditional neural networks mainly face structured data, while many biomedical data are unstructured data. The recent progress of graph model deserves high attention. 

Gao Lin made a keynote speech entitled "key problems and challenges driven by single cell data", pointed out that the breakthrough progress of single cell omics technology has brought biomedical research into the single cell level, and emphatically discussed the challenges faced by single cell data calculation and analysis from the perspectives of cell type definition and cell interaction. Shen Hongbin made a keynote speech entitled "discussion on biomedical big data and artificial intelligence research", pointed out that deep neural network plays an increasingly important role in medical image and omics data processing, and put forward four challenges: multi-omics and multi-modal data processing, data processing with serious batch effect, construction of stability and interpretability model The challenge of algorithm and computing power.

In the problem discussion session, the guests focused on "what are the special theoretical and methodological challenges of artificial intelligence for biomedical big data?" "What is the most likely breakthrough in the near future about the combination of biomedical big data and artificial intelligence?" "What are the prospects for the industrialization of biomedical big data and artificial intelligence? IBM and Google have successively failed in the field of health. What are the difficulties and bottlenecks in the industrialization of this field?" Three issues were discussed interactively.

This activity used the mode of online thematic discussion, which aimed to discuss the research status and future development trend of the frontier intersection of biomedical big data and artificial intelligence. More than 100 members of CAAI bioinformatics and artificial life special committee and CAA intelligent health and bioinformatics special committee attended the meeting online, and more than 200 people watched the live broadcast online.

This paper is contributed by CAAI bioinformatics and artificial life special committee