Jianfeng's Homepage

                 

 

[CV] --[Teaching] --[Publications]

----------------------------------------------------------------------

Research Interests

Based upon multi-scale neuroscience, biological and clinical data, our group are working on developing novel machine learning algorithms to tackle challenging issues raised in brain disorders: identify biomarkers for both disease diagnosis and prediction in dementia, Parkinson's disease, depression, schizophrenia and addiction etc. Our studies often lead to breakthroughs and successful applications in clinical practice, as partially reflected in many publications in Nature-related journals and some of our spin-off companies. Equipped with all the novel knowledge from our computational neuroscience and related studies as described above, we are then working on developing the digital twin model of the whole brain, from human, monkey, rat to zebrafish. This approach in turn enables us to explore the fundamental issues of brain-inspired AI, an alternative way to achieve artificial general intelligence, echoed many years ago by Richard Feynman "What I cannot create, I do not understand". Often funding is available for PhD, Postdocs and young faculty member positions. We invite prospective PhD applicants to contact us for more information.

1. Healthy/Diseased brain

We have accumulated some of the largest data sets in the world for the human brain, ranging from behavioural, symptomatic data to genetic and proteomic data. Interested in seeing these datasets or want to work with us? Please do contact me via email. Some recent publications include studies on: Depression; Nature Mental Health (2023) [1] ; Schizophrenia: Nature Mental Health (2023) [2] ; Addiction: Nature Communications (2023) [3] ; Dementia: Nature Aging (2024) [4] , also in Nature News .

2. Whole brain simulation

We have used one of the world's largest HPCs to carry out a simulation of 86B neurons which maps to the scale of the human brain (digital twin brain (DTB)). The human DTB has a correlation coefficient of 0.9 with its biological counterpart in resting-state
Human DTB [5] [6];     Zebrafish DTB (with Dr. Du JL) (see Fig. 1). :

3. Machine learning

We are working on analyzing and developing various algorithms such as supervised learning, unsupervised learning and community detection algorithms with applications. Our spinoff companies   NeuroBlem;       Automatic car  

       

Fig. 1: Our digital twin brain of Feng's brain.