How are NIBIB-funded researchers using computational modeling to improve health?
Modeling infectious disease spread to identify effective interventions. Modeling infectious diseases accurately relies on numerous large sets of data. For example, evaluation of the efficacy of social distancing on the spread of flu-like illness must include information on friendships and interactions of individuals, as well as standard biometric and demographic data. NIBIB-funded researchers are developing new computational tools that can incorporate newly available data sets into models designed to identify the best courses of action and the most effective interventions during pandemic spread of infectious disease and other public health emergencies.
Tracking viral evolution during spread of infectious disease. RNA viruses such as HIV, hepatitis B, and coronavirus continually mutate to develop drug resistance, escape immune response, and establish new infections. Samples of sequenced pathogens from thousands of infected individuals can be used to identify millions of evolving viral variants. NIBIB-funded researchers are creating computational tools to incorporate this important data into infectious disease analysis by health care professionals. The new tools will be created in partnership with the CDC and made available online to researchers and health care workers. The project will enhance worldwide disease surveillance and treatment and enable development of more effective disease eradication strategies.
Transforming wireless health data into improved health and healthcare. Health monitoring devices at hospitals, and wearable sensors such as smartwatches generate vast amounts of health data in real-time. Data-driven medical care promises to be fast, accurate, and less expensive, but the continual data streams currently overwhelm the ability to use the information. NIBIB-funded researchers are developing computational models that convert streaming health data into a useful form. The new models will provide real-time physiological monitoring for clinical decision making at the Nationwide Children’s Hospital. A team of mathematicians, biomedical informaticians, and hospital staff will generate publicly shared data and software. The project will leverage the $11 billion wireless health market to significantly improve healthcare.
Human and machine learning for customized control of assistive robots. The more severe a person’s motor impairment, the more challenging it is to operate assistive machines such as powered wheelchairs and robotic arms. Available controls such as sip-and-puff devices are not adequate for persons with severe paralysis. NIBIB-funded researchers are engineering a system to enable people with tetraplegia to control a robotic arm while promoting exercise and maintenance of residual motor skills. The technology uses body-machine interfaces that respond to minimal movement in limbs, head, tongue, shoulders, and eyes. Initially, when the user moves, machine learning augments the signal to perform a task with a robotic arm. Help is scaled back as the machine transfers control to the progressively skilled user. The approach aims to empower people with severe paralysis and provide an interface to safely learn to control robotic assistants.