IdentiARAT: Toward Automated Identification of Individual ARAT Items from Wearable Sensors
This addresses the time-consuming and subjective manual assessment by clinical staff, though it is incremental as it builds on existing sensor and classification methods.
The study tackled the problem of automating the labeling of ARAT items for upper limb motor function assessment by using wrist-worn inertial sensors and MiniROCKET classification, achieving reliable classification of ARAT domains but with challenges in distinguishing similar items.
This study explores the potential of using wrist-worn inertial sensors to automate the labeling of ARAT (Action Research Arm Test) items. While the ARAT is commonly used to assess upper limb motor function, its limitations include subjectivity and time consumption of clinical staff. By using IMU (Inertial Measurement Unit) sensors and MiniROCKET as a time series classification technique, this investigation aims to classify ARAT items based on sensor recordings. We test common preprocessing strategies to efficiently leverage included information in the data. Afterward, we use the best preprocessing to improve the classification. The dataset includes recordings of 45 participants performing various ARAT items. Results show that MiniROCKET offers a fast and reliable approach for classifying ARAT domains, although challenges remain in distinguishing between individual resembling items. Future work may involve improving classification through more advanced machine-learning models and data enhancements.