Performance Gains Through Sensory Systems

Over recent years, sensors have increasingly become used to improve the performance of humans. Popular sensors provide a cost efficient way to gather various inputs ranging from temperature to movements to sound. Today, smartphones are packed full of sensors that enable us to go about our daily lives allowing us to find the closest restaurant and get turn-by-turn directions at a glance. External Bluetooth sensors are also integrated to help aid in medical tasks such as checking glucose levels or monitoring weight. These sensors have been so ingrained in everyday living that it is hard to imagine life before their existence. In fact, a good amount of our performance and decision making process relies on the information we gather from these sensors. Two main demographics, in particular, benefit from performance improvement sensors. The first demographic is older adults. Several sensor-based systems have been created to help older adults perform at a higher level, which increases their quality of life. Fall monitor systems are being created using various sensors such as accelerometers, video cameras, and acoustic sensors. GPS sensors are being used to create wandering tracking systems of dementia patients. Various other systems have also been constructed to assist with the day-to-day medical care of older adults. While targeted for different purposes, they all have the same goal, which is to positively increase the performance of the user. The other demographic that sees a marked performance increase is athletes. In general, a key difference between older adults and athletes performance level. Older adults may display minimal function while athletes may display advanced function. There have been several approaches that offer ways of improving the performance for both demographics. For older adults, systems are available that allow them to live more independently and provide peace of mind to loved ones. The systems achieve this goal by using sensors to monitor the user and automatically send alerts in an emergency. These emergencies can range from falling and not being able to get up or wandering outside in extreme conditions and becoming lost. On the other hand, other systems use sensors to evaluate and train athletes at the highest level. Often times, these systems are designed with speed and information as a key goal. They aim to improve several functions such as reaction time, spatial awareness, and agility. Data from the sensors is commonly evaluated in order to fine tune the athlete's movements that may be sport specific. While sensors provide valuable information, they can be limited in several ways. One main concern is erroneous output from excessive noise. In the case of purely vision-based systems, background objects and movements create unwanted data that must be filtered. On the other hand, systems based only on inertial sensors incur noise when mounted to body parts that frequently move, such as hands. In addition to noise, single sensor systems are limited by processing time. Most video capturing inputs and processing algorithms are capable of running at 60 frames per second or every 1000 milliseconds. However, reaction time occurs on the order of 50-100 milliseconds, which will require additional time to compute (or expensive specialized hardware). One way of addressing this issue is by using several sensors. Fusing the inputs from several sensors provides a robust, context-rich collection of data. This data can be used in numerous applications to better the fields of medicine, sports, and computer science. One particular area that can benefit from such sensor-fused systems is the improvement of visual cognition. Visual cognition is the process of decoding information visually as it is collected by the eyes and moves into the brain's waves. These brain waves then perform object recognition and invoke memories and emotions. With assistance from these sensory systems, people can be trained to see better and faster while strengthening the neural connections in the brain. This dissertation explores a training program aimed to improve the visual performance of athletes. The training program consists of several exercises designed to workout the visual system of the trainee. Both commercial and custom sensors are used to gather data and evaluate the progression of athlete through the program with special focus on reaction time and visual evoked potentials. Several algorithms are implemented to evaluate the data and a novel sensor fused, reaction time algorithm is proposed.