Human Computer Interaction (HCI) is central for many applications, including hazardous environment inspection and telemedicine. Whereas traditional methods ofHCI for teleoperating electromechanical systems include joysticks, levers, or buttons, our research focuses on using electromyography (EMG) signals to improve intuition and response time. An important challenge is to accurately and efficiently extract and map EMG signals to known position for real-time control. In this preliminary work, we compare the accuracy and real-time performance of several machine-learning techniques for recognizing specific arm positions. We present results from offline analysis, as well as end-to-end operation using a robotic arm.

Sleep apnea is characterized by a cessation in breathing during periods of sleep. Sleep apnea affects over 46 million Americans, with 80% of them undiagnosed. Diagnosis methods are expensive and uncomfortable. Sleep assessment methods can be performed either at a sleep lab in a clinic or at home. In-lab sleep tests, called polysomnograms, are considered uncomfortable due to not sleeping in one's customary environment, and having over 10 electrodes or attachments touching the body on the head, nasal cannula, and limbs. At-home tests may be hard to set up correctly, and some have many wires and attachments that reduce comfort. We are developing a simpler sleep apnea recording system that restricts the electrodes and sensors to the area of a shirt. Electrocadiogram signals are recorded by three electrodes, and electromyogram of the intercostal muscles are recorded to monitor respiratory activity. A prototype has been made from an athletic shirt which houses all electronics. Custom made LabView programs analyze and plot the signals for assessment of variability of the heart rate and respiration rate, which may be used to identify possible episodes of sleep apnea. Further development and testing are necessary toward wider application.

Certain cancer patients benefit from chemoradiotherapy that simultaneously applies radiation therapy and chemotherapy to targeted tissue. Infusion pumps administer the chemical agents for chemotherapy. Inadvertently during the treatment, radiation penetrating into the infusion pumps may damage the internal solid state, integrated circuit chips. Such defects may result in inaccurate dosing. Although the manuals for infusion pumps from the manufacturers warn against device use in radiological environments, the devices continue to be used during chemoradiotherapy. Proper shielding of the devices during exposure to radiation could minimize risk of device malfunction. Lead shielding has disadvantages of toxicity, weight, and bulkiness. In this study we have researched and tested potential non-lead composites for shielding. Micro particles of Bismuth (III) oxide (BO) were impregnated into two materials at various percent weights: Polydimethylsiloxane (PDMS) and Polyurethane Elastomer (PU). PDMS/BO samples attenuated X-rays of 52 kVp at 4 mm thickness. PU/BO attenuated X-rays of 52kVp at 3 mm thickness. The addition of BO to industry plastics shows promise toward use as shield to reduce risk of X-ray induced defects for infusion pumps within diagnostic ranges of radiation (1-150kVp). More development and testing are necessary to explore the effectiveness of the material at therapeutic levels (>150kVp) and to expand the potential clinical applications.