主 题: Peer into Your Brain via Contiguous-scale Data
报告人: Prof. Jianfeng Feng (Shanghai Center of Mathematical Science, Fudan University)
时 间: 2018-04-20 10:00-11:00
地 点: Room 1114, Sciences Building No. 1
报告摘要：With the available data of huge samples of contiguous-scales both for healthy controls and patients including depression, autism and schizophrenia etc, we are in the position to quantify human brain activities such as creativity, happiness, IQ and EQ etc and search the roots of various mental disorders. With novel mathematical and machine learning approaches, we first introduced functional entropy and entropy rate of resting state to characterize the dynamic behaivour of our brain. It is further found that the functional entropy is an increasing function of age, but a decreasing function of creativity and IQ. Its biological mechanisms are explored. With the brain wide associate study approach, for the first time in the literature we are able to identify the roots of a few mental disorders. For example, for depression, we found that the most altered regions are located in the lateral and medial orbitofrontal cortex for punishment and reward. Follow-up rTMS at the lateral orbitofrontal cortex demonstrated significant outcomes of the treatments. Finally we discuss some of our recent results on brain-inspired AI and their applications.
报告人简介：冯建峰，国家第二批“千人计划”特聘教授、上海数学中心首席教授、复旦大学类脑智能科学与技术研究院院长。长期致力于发展数学、统计与计算机的理论和方法，并原创性地将它们应用于解决神经科学,各类脑疾病和智能算法中的具体问题。作为英国已做出突出贡献和具有巨大潜力的科学家，2011年获英国皇家学会的沃夫森研究功勋奖（首位华人）。在单神经元和神经元网络的动力学研究、机器学习算法的设计和分析、随机控制理论、因果关系分析等方面都做出过杰出工作，发表在Molecular Psychiatry, Brain, PNAS, PRL, J Neuroscience, IEEE 系列上。目前研究兴趣主要集中在不同尺度的海量数据的分析，挖掘和理论研究上。提出和发展了全脑关联分析（BWAS）的方法和理论，并成功的应用于抑郁症，精分和自闭症的病灶的发现；发展了定量化大脑的泛函熵方法，并成功的应用于老龄化，智商和创造性的研究中。
Jianfeng Feng is a thousand-talent program (second round) professor, the chair professor of Shanghai National Centre for Mathematic Sciences, and the Dean of Brain-inspired AI Institute in Fudan University. He has been developing new mathematical, statistical and computational theories and methods to meet the challenges raised in neuroscience and mental health researches. Recently, his research interests are mainly in big data analysis and mining for neuroscience and brain diseases. He was awarded the Royal Society Wolfson Research Merit Award in 2011, as a scientist ‘being of great achievements or potentials’. He has made considerable contributions on modelling single neurons and neuronal networks, machine learning, and causality analysis with publications on Molecular Psychiatry, Brain, PNAS, PRL, J Neuroscience etc. He has proposed and developed BWAS method (Brain-wide association study), and successfully applied it to search the roots in depression, schizophrenia and autism; developed functional entropy method and applied it to the study of ageing, IQ and creativity etc.
