Developing and validating an accelerometer-based algorithm with machine learning to classify physical activity after acquired brain injury

Purpose: To develop and validate an accelerometer-based algorithm classifying physical activity in people with acquired brain injury (ABI) a laboratory setting resembling real home environment.Materials methods: A development validation study was performed. Eleven healthy participants 25 patients ABI performed protocol of transfers ambulating activities. Activity measurements were accelerometers thermal video camera as gold standard reference. machine learning-based specific activities from the accelerometer data developed cross-validated training sample 11 participants. Criterion validity established 3 models same ABI.Results: Modeled on 15 ABI, had good precision for 10 ABI. The weighted sensitivity all 89.3% (88.3–90.4%) positive predictive value 89.7% (88.7–90.7%). differentiated between lying sitting activities.Conclusion: An to classify populations its criterion established. Further testing settings continuous monitoring is warranted.