- Apply to UW
- Programs & Majors
- Cost & Financial Aid
- Current Students
- UW Life
- About UW
Dr. Novak’s group conducts research on diverse aspects of rehabilitation engineering and human-machine interaction in general. We mostly focus on software development and human subjects evaluations, with occasional forays into electronic and mechatronic development. Some of the group’s recent projects are presented below.
Low back pain is a leading cause of disability worldwide, with lifetime prevalence estimates as high as 84%. In the United States, over 8 million people have only a limited ability to perform everyday activities due to low back pain, resulting in 149 million work days lost per year and an annual cost of over 100 billion dollars. As pharmaceutical and surgical methods for low back pain management are only recommended for severe cases, there is a great need for noninvasive technical solutions that could effectively prevent or relieve back pain.
Together with UW's Division of Kinesiology and Health and Livity LLC (a Colorado-based startup), Dr. Novak’s research group is working on trunk exoskeletons that can dynamically support the trunk, relieving the load on different muscles. While the first developed exoskeleton was purely passive, a recent grant from the National Science Foundation's Mind, Machine and Motor Nexus will support the development of exoskeletons that can sense the user's motion intentions and dynamically change their configuration in order to more efficiently support the user, hopefully resulting in greater health benefits. The project involves cutting-edge research in sensor fusion, mechatronics, and biomechanics.
Recent publications about trunk exoskeletons:
A trunk exoskeleton worn by a study participant.
Testing the exoskeleton’s effect on stability in response to external perturbations.
Can my computer understand my feelings? Though the question may sound silly, it is the focus of the emerging field of affective computing, which tries to recognize human emotions and respond to them. This is done through measurements of heart rate, sweat, brain activity and other physiological responses. From these measurements, it is possible to infer that, for example, increased sweating and heart rate may indicate stress. Such affective computing has numerous applications, such as intelligent cars that detect when the driver is drowsy or office computers that detect when a worker is too stressed to work effectively.
Our group is exploring both fundamental and applied issues of this technology, hoping to move it from laboratory prototypes to practically usable and beneficial applications. Together with UW’s Department of Psychology, we have been funded by a grant from the National Science Foundation’s Division of Information and Intelligent Systems to develop new pattern recognition and machine learning methods for affective computing. Furthermore, we previously received applied grants to develop effective computing techniques for driver drowsiness prevention together with UW’s Department of Civil and Architectural Engineering.
Recent publications about affective computing:
A participant driving in the WYOSIM driving simulator while their attention levels are monitored with physiological sensors.
Despite great advances in medicine, humanity still faces countless diseases. Not all of these diseases are fatal; many leave the victim alive, but with permanently impaired motor and cognitive abilities. For example, stroke has a survival rate of approximately 70%, but the vast majority of survivors are left with long-lasting motor disorders. Intensive post-stroke exercise increases the chance of full recovery, but most stroke victims do not receive adequate exercise due to a shortage of qualified medical staff. Therefore, there is a great need for rehabilitation technology that would support and complement therapists.
Many devices have been developed to support motor rehabilitation, ranging from simple motion sensors to powered exoskeletons. Our research group has worked on a novel approach: devices that allow patients to compete or cooperate with either other patients or unimpaired loved ones. This has two potential benefits. First, competition and cooperation can result in greater patient motivation, ensuring longer and more intense exercise. Second, they can increase the quality of motor learning, ensuring that patients relearn motions more quickly and effectively. Our research on the topic is highly interdisciplinary, combining engineering, computer science, psychology and kinesiology.
Recent publications about collaborative and competitive exercises:
Two participants exercising with a collaborative virtual environment and two Bimeo arm rehabilitation devices.