BRAIN-MACHINE INTERFACES
Assistive devices that people control using only their minds
Caltech neuroscientist Richard Andersen has collaborated with clinicians to develop brain-machine interfaces (BMIs) that help people direct prosthetics and computer cursors using only their intentions. However, BMIs developed to date are bulky and require electrodes that are not stable over time.
Now, Andersen and Azita Emami, an expert in wireless biomedical devices, are codesigning tiny, implantable chips and machine-learning algorithms that will record, decode, and communicate neural information with extreme efficiency.
It is amazing to see tetraplegic patients control robotic limbs and cursors on the screen in clinical settings with current BMIs. But imagine if a tiny, implantable device could empower patients to move mind-controlled wheelchairs and to perform similar tasks wherever they go.
NONINVASIVE SENSORS
Noninvasive sensors that can monitor conditions ranging from stress and depression to cardiovascular and kidney disease
Caltech professor Wei Gao has invented a mass-producible, wireless sensor that uses sweat to accurately detect levels of cortisol, the body's stress hormone. The sensor is paired with integrated circuits that independently analyze and transmit results to patients' physicians via smart phones.
This marks a significant improvement over current tools to diagnose stress and depression, which rely on subjective questionnaires or blood tests that must be evaluated by trained professionals. What is more, the tests themselves can induce stress that skews results.
Gao also is developing e-skin, which generates energy from sweat, and another sensor that measures uric acid levels to monitor cardiovascular conditions, diabetes, or kidney disease.
Investigator
Wei Gao, Assistant Professor of Medical Engineering
Wearable sweat sensors have the potential to rapidly, continuously, and noninvasively capture changes in health at molecular levels. They could enable personalized monitoring, early diagnosis, and timely intervention.
BREAST-CANCER SCREENING
Fast, accurate, and painless breast-cancer screening
Early detection increases breast cancer survival rates. However, many women avoid mammograms because they hurt, can yield false positives, and come with a dose of X-ray radiation.
A new option, called photoacoustic computed tomography, shines pulses of light into the breast and can find tumors in seconds. The diffused light causes hemoglobin in red blood cells to vibrate as an array of 512 sensors placed on the skin detects the vibrations. In turn, algorithms process these detections to image structures inside the breast.
Investigator
Lihong Wang, Bren Professor of Medical Engineering and Electrical Engineering
This is the only single-breath-hold technology that gives us high-contrast, high-resolution, 3-D images of the entire breast. Our goal is to build a dream machine for breast screening, diagnosis, monitoring, and prognosis [that is] fast, painless, safe, and inexpensive.