LUCID CIRCUIT, INC. — Department of Health and Human Services SBIR Phase I: 600

Description

Abstract Medical decision making in intensive care settings relies on complex biomedical informationcollected in real timeto assist hospital staff in patient management and intervention decisionsIn this contextmachine learningMLapplications applied to healthcare analytics have been shown to provide significant improvement in quality of care and overall efficiency in ICU operationsA number of challenges affect healthcare analyticsthese are growing data sizesincreasingly heterogeneous data sourcesdata assurancecredibility and quality as well as the privacy of patient record interchangeCurrent approaches to healthcare analytics are predominantly reliant on desktop and cloud computing architecturesThese traditional systems exhibit high latencyhigh power requirementsand are overall inefficient for advanced analytic applicationsIf we hope to leverage the benefits of growing data size and heterogeneitya new hardware paradigm for real timesecure and reliable ML in the clinic is neededA potential for great innovation in medical big data applications is to imbue more analytic capability into smaller devices that sit closer to the source of dataThis so called edge computing will be instrumental in future medical applicationswhere real timelow latencylow power computational tools are likely to dramatically improve the feasibility of remote monitoring and intensive monitoring applicationsultimately improving health outcomesLucid CircuitIncis developing a line of purpose built edge analytics processors called AstrumTMAstrumTM is an energy efficient runtime adaptable processor for reliable high performance computing and low power applicationsAstrumTM processors benefit from an adaptable compute fabric that combines runtime reconfigurable architecture and in silicon security featuresAstrumTM processors are designed to supports evolving analytics algorithmsIn this proposalwe seek to optimize state of the art ML algorithms for signal processingdata quality monitoring and safe data interchange in remote and intensive monitoring medical applications at the edge using AstrumTMIn phase IIwe seek to develop an analytics programming ecosystem within which developers can customize signal quality metrics and prediction rules at a high level of abstractionusing popular analytics programming platformsUltimatelythe work proposed in this application will enable the development of a new generation of edge based precision intensive care and mobile health devicestogether with a programming toolkit for customizationtuningand development of ML algorithms that best match the needs of researcherspatientsand healthcare providers Project Narrative Machine LearningMLapplications to medical big datahave been shown to provide improved quality of care in intensive care settingsTheir implementation in the clinic is currently limited by reliance on obsolete desktop and cloud computing architecturesBy leveraging our experience with purpose built AstrumTM processorswe propose to adapt state of the art analytics for computation at the edgeaiming to deliver edge based precision intensive care analytics capabilities to a new generation of intensive care and mobile health devices