BACKGROUND
Tourette syndrome is a neurological disorder that is characterized by repeated unintentional physical movement and vocal sounds, better known as tics. Cases of mild Tourette can have tics numerous times throughout the day, while severe cases may have tics every 5 to 10 seconds. At certain times, typically during high levels of stress, tics become chained in an incessant, continuous fashion—this is known as a tic attack. Tic attacks incapacitate the patient, rendering it difficult for them to move, perform daily actions, and even communicate with others. Caretakers—usually guardians, family members, or nurses—can help reduce the time tic attacks last with their presence and by providing emotional support to the patient.
OBJECTIVE
We describe TSBand, a wearable wristband that uses machine learning algorithms and a variety of sensors to monitor for tic attacks and notify caretakers when an attack occurs.
METHODS
We conducted a research survey with 70 Tourette patients to determine the usability and functionality of TSBand; internal review board approval was not required.
RESULTS
This study has resulted in a smart wristband prototype that costs US $62.74; it uses movement, heart rate, sweat, and body temperature to detect tic attacks using a hybrid local outlier factoring and regression algorithm. An audio tic attack detection mechanism is also included, using recurrent neural networks, and a manually activated backup button and backup audio mechanism are fitted to alert caretakers on the personalized companion app.
CONCLUSIONS
TSBand enables the caretaker to provide support faster and prevent excessive self-harm or injury during the attack. It is an affordable and effective solution, solving a problem that many Tourette patients, often children, face. This study has not had the opportunity to test TSBand with any Tourette patients, and we aim to perform rigorous testing and analysis after grant funding is secured.