MedTech Speaker Series, July 2019 - Xilinx and Camensys

Event Title: MedTech Speaker Series, July 2019 - Xilinx and Camensys

Date: Wed, 07/17/19

Time: 6:00 pm - 8:30 pm

Venue: Quinlan Community Center, Cupertino Hall

10185 N Stelling Rd, Cupertino, CA 95014

Thanks to sponsorship of this event by Xilinx, please register free of cost at https://www.med-tech.org/events as we have limited seating capacity. Please bring your registration confirmation email to the event as the entry pass.

Who should attend this event:

If you are involved as product manager, product designer, architect, developer, quality manager or regulatory affair manager in developing Medical Device Hardware + Software or Software as Medical Device (SaMD) that uses AI or Machine Learning you should attend this event. The takeaways from this event will include:

• Information about industry leading FPGA based implementations for AI/ ML algorithms that beat CPUs/GPUs to provide low latency and low power solutions. This has particular relevance to software used for medical devices.
• Whether you are in early stage in your product development or close to submitting for FDA approval, you will learn various ISO standards and FDA guidelines that you should implement for the Medical Device Software or Software as Medical Device (SaMD) especially that uses AI or Machine Learning. This includes quality and design controls which are relevant for all involved in the product development.

Session 1: Turbocharge AI Based Applications like Medical Device Software using Reconfigurable Platforms

Speaker: Ashish Sirasao, Fellow AI Solutions at Xilinx

Summary: Artificial Intelligence (AI) algorithms including Deep learning algorithms and their combinations are now being used to solve problems in new generation applications of many areas, and in particular areas that have large data sets and complex prediction models like the medical space. With the increasing scale of deployment, efficient computational hardware has become critical for large-scale implementation and deployment of these algorithms. Xilinx reconfigurable platforms enable optimization of AI using variable precision arithmetic, customized memory hierarchy, and model compression to address broader figures of merits, like throughput, latency, energy, and accuracy. Xilinx architectures are extremely suited for applications requiring low latency and low power required in medical applications. In this presentation, we present how reconfigurable platforms can enable optimized hardware and software co-design methodology for efficient implementation of a wide range of AI algorithms especially related to medical applications.

Session 2: Are you ready for FDA with your Machine Learning based SaMD

Speaker: Sky Basu, Chief Consultant and CEO, Camensys

Summary: With the new generation of Software as Medical Device (SaMD) with Machine Learning as one of the core technologies, FDA has shared some early guidance on how they would like to regulate these learning devices. Good Machine Learning Practice (GMLP) coupled with Total Product Life Cycle (TPLC) can create a solid foundation of Culture of Quality and Organizational Excellence (CQOE). This presentation will explain the lay of the land of quality and compliance requirements covering the above alphabet soup for developers of SaMD with Machine Learning.

6:00pm: Door opens

6:00pm – 6:30pm: Networking and refreshment

6:30pm – 8:30pm: Presentation Sessions with Q&A