A Classifier Algorithm to Identify 30-second Sit-to-Stand Tests in Patients with Multiple Sclerosis
Abstract
Multiple sclerosis (MS) is a neurological condition that causes progressive and often irreversible damage to the central nervous system. Many symptoms of MS impair patients’ mobility, leading to a greater incidence of falls. The 30-second sit-to-stand test (30CST) is a common assessment of balance and strength in MS patients, and one metric wherein differences in accelerometer data between fallers and non-fallers could lead to accurate prediction of fall risk in the future. 40 subjects (20 fallers, 20 non-fallers) will wear accelerometers on their chest, right thigh, and right foot during normal activity for 36 hours. Each patient will conduct a 30CST eight times a day. In order to expediently analyze 30CST occurrences from patients’ at-home accelerometer data, a classifier algorithm that can automatically identify those tests will be developed.
A Classifier Algorithm to Identify 30-second Sit-to-Stand Tests in Patients with Multiple Sclerosis
Multiple sclerosis (MS) is a neurological condition that causes progressive and often irreversible damage to the central nervous system. Many symptoms of MS impair patients’ mobility, leading to a greater incidence of falls. The 30-second sit-to-stand test (30CST) is a common assessment of balance and strength in MS patients, and one metric wherein differences in accelerometer data between fallers and non-fallers could lead to accurate prediction of fall risk in the future. 40 subjects (20 fallers, 20 non-fallers) will wear accelerometers on their chest, right thigh, and right foot during normal activity for 36 hours. Each patient will conduct a 30CST eight times a day. In order to expediently analyze 30CST occurrences from patients’ at-home accelerometer data, a classifier algorithm that can automatically identify those tests will be developed.