行健讲坛学术讲座
第379期
时间: 2019年4月10日（周三）上午9：30
所在: 校本部东区翔英大楼 T516室
讲座: 基于人工智能的康健监护：从音频到自觉身体活动领域的应用
A.I.4Healthcare: Applications from Audio to Spontaneous Physical Activity
演讲者: 钱昆 特任研究员 日本东京大学
演讲者简介：钱昆博士，， ，，日本学术振兴会特殊研究员（全球每年任命率10%），， ，，日本东京大学特任研究员，， ，，德国慕尼黑工业大学工学博士，， ，，主要研究人工智能与信号处置惩罚偏向。。。。。钱博士与美国卡内基梅隆大学、英国帝国理工大学、新加坡南洋理工大学、中国科学院等海内外顶尖高校和科研机构坚持相助关系，， ，，致力于深度学习在医疗康健、音频智能感知和数据挖掘方面的研究。。。。。现在以第一作者身份揭晓SCI收录期刊论文8篇，， ，，其中包括IEEE Transactions on Biomedical Engineering、Annals of Biomedical Engineering、JASA等国际着名期刊，， ，，并在国际着名学术聚会如ICASSP、EMBC、GlobalSIP上揭晓相关学术论文。。。。。钱博士恒久担当国际着名期刊IEEE Transactions on Cybernetics、IEEE Transactions on Biomedical Engineering、IEEE Transactions on Affective Computing、IEEE Transactions on Neural Networks and Learning Systems、IEEE Signal Processing Letters以及ICASSP、INTERSPEECH、EMBC、AVEC等国际着名学术聚会审稿人。。。。。已授权中国发明类专利3项，， ，，德国专利1项。。。。。
讲座摘要：In a traditional or classical A.I. paradigm, the human hand-crafted features are extracted from the data by several signal processing methods, e.g., Fourier transformation, wavelet transformation, empirical mode decomposition, etc. Subsequently, a machine learning model can be trained when fed with those features. Even though the performance and the robustness of the model could be feasible for further implementations in real practice, the feature engineering process, which needs specific domain knowledge, is still time-consuming, and expensive. As an emerging technique, deep learning, can make it possible to make models learn higher representations from the data itself. In this presentation, Dr. Qian will present his main work in Technical University of Munich, Germany, and his most recent work in The University of Tokyo. For his work in Germany, the audio data can be used for diagnosing some diseases related to the knowledge of body acoustics. For his work in Japan, the spontaneous physical activity data can be good representations for screening the patients suffering from the major depressive disorder.
约请人：8188cc威尼斯通讯与信息工程学院 朱梦尧副教授
接待宽大西席和学生加入！
