Office of Technology Transfer – University of Michigan

Human Behaviorome Capture and Recognition for Disease Phenotyping and Healthcare Improvement

Technology #5775

Questions about this technology? Ask a Technology Manager

Download Printable PDF

Fan Meng
Managed By
Drew Bennett
Associate Director - Software Licensing 734-615-4004

This is an activity recognition technology used for continuous monitoring and detection of specific behaviors of people in their natural environments. It can be used for monitoring the compliance of any medication regimen, food intake, exercise, and other specified behaviors. According to the New England Healthcare Institute, medication adherence presents a huge problem that leads to poorer health, more frequent hospitalization, and $290 billion in increased annual medical costs. As many as half of all patients in the US do not take their medications as instructed, including failing to take medication on time or in the proper dose. Patients with chronic conditions such as diabetes and high blood pressure are among the groups that are less likely to follow their medication regimen. As remote patient monitoring and wearable sensor technology continue to improve and expand into the market, they present a unique opportunity to address the problem of patient compliance, improve outcomes, and decrease healthcare costs.

Behavior Monitoring Using Wireless Technology

Researchers in the Department of Psychiatry at the University of Michigan propose using a combination of a wireless sensor with a smartphone to monitor specific human behaviors. The activity sensor is worn on the part of the body that is expected to participate in the activity. It is calibrated to a particular motion, for example, opening of a bottle of medication and taking the medicine, based on the patient’s own movements. The data generated by the sensor is wirelessly relayed to the smartphone, analyzed, and an automatic response is generated. If a patient does not take medication during a set time frame, a reminder alert is sent. A time stamp is generated for each time that a patient takes the medicine, and this information is stored and can be accessed by a healthcare professional in real-time for immediate analysis. Given the widespread use of smartphones and the low cost of wireless-enabled communication technologies, this technology can be quickly adopted by large populations.

Applications • Remote patient monitoring for medication compliance • Assessment of drug efficacy based on behavioral manifestations • Real-time alerts and interventions • Identification of behavioral biomarkers

Advantages • High resolution and accuracy of physical activity identification • Adaptable to a wide variety of situations • Low cost and fast implementation