报告名称 : Trade-offs in Algorithm and Hardware Design
报告时间 : 2019年7月8日 16：00
报告地点 : 主楼四区209
报 告 人 : Robert Karam教授 南佛罗里达大学
个人简介
Dr.Karam is an assistant professor in the department of computer science andengineering at the University of South Florida. He received the BSE. and MSdegrees in computer engineering from Case Western Reserve University in 2012and 2016, and the PhD degree in computer engineering from the University ofFlorida in 2017. From 2012 to 2017, he was a biomedical engineer in theAdvanced Platform Technology Center, L. Stokes Cleveland VA Hospital. Dr. Karamhas published over 30 articles and abstracts in top journals and conferences.He is the recipient of the National Science Foundation Award for YoungProfessionals Contributing to Smart and Connected Health, and the Best PaperAward from the 2016 IEEE Biomedical Circuits and Systems conference. Hisresearch interests include reconfigurable hardware security, and algorithms andhardware design for biosignal processing in implantable electronics. He is amember of the IEEE and ACM.
报告摘要
In the era of digital healthcare,implantable electronics have emerged as promising diagnostic, monitoring, andtreatment tools for improving patient outcomes. On-chip processing of biomedicalsignals is becoming ever more common due to improvements in integration densityand low power design techniques, enabling us to perform increasingly complextasks in even highly constrained devices. Complex on-chip signal processingalso opens the door to novel treatment options, such as closed-loopneuromodulation. However, battery life and reliability are still majorconcerns, necessitating careful algorithm design for biosignal processing thatnot only meets the needs of doctors and patients, but is amenable toimplementation in ultra-low power and area-constrained devices. This seminarwill give an overview of the challenges in algorithm and hardware design fortunable signal compression and feature extraction in a wireless, implantablepressure monitor, and explore future directions for the research.
