STTR Phase I: Developing a Machine Learning Platform for Workplace Wellness Programs

点击翻译按钮获取中文摘要

The broader impact/commercial potential of this Small Business Technology Transfer (STTR) Phase I project promotes a promising new approach to improve employee wellness with the potential for wider application beyond the workplace. This improved approach combines both innovative clinical diagnostics with computer assisted learning to guide recommendations. It offers the potential to curb epidemic chronic disease through the use of efficient as well as scalable operating models. Adopting this data-driven approach is expected to improve not only participating employees? health but also reduce health care costs. Increased productivity which follows from improved employee health can also be expected. Accordingly, better health and economic savings to both corporate America and working Americans are realistic objectives of this project. Employers and the general public therefore are expected to benefit from this initiative.The proposed project combines novel as well as economic use of clinical diagnostics, strengthened by computer learning, to effectively improve workplace wellness. Conventional wellness strategies are presently both inefficient as well as ineffective in addressing the growing epidemic of chronic diseases. These chronic diseases include but are not limited to: diabetes, obesity, high blood pressure and cardiovascular diseases. The principal objective of this project is the integration of various clinical findings in prevention of chronic disease enhanced by machine learning. This unique approach aids in the development of scalable, workplace wellness programs which are effective and affordable. The novel platform continuously learns from all participants? interventional results and intelligently discovers the best preventive intervention plan for each participant. The platform innovatively adopts low cost quantitative diagnostic imagery to track key health parameters clinically correlated to wellness. Computer aided modeling includes dynamic physiological models (i.e. current symptoms, lifestyle measurements, past medical history, MRI scans and laboratory tests) to clinically evaluate the course of chronic disease along with various prevention strategies for intervention. Measures of clinical outcome enhanced by machine learning identify important biomarkers to develop intervention recommendations for participating individuals.