Home-Based Training With Feedback to Improve Outcomes in Adolescents and Young Adults With Cerebral Palsy..

Movement data will be collected using seven Inertial Measurement Units (IMUs) worn continuously for 72 hours at three time points: baseline, 12 weeks (end of intervention), and 24 weeks (follow-up).

IMUs will be attached to the sternum, wrists, thighs, and lower legs using adhesive patches and will capture 3-axis accelerometer and gyroscope data. A validated custom neural network (Novosel et al. 2023) will convert signals into images and analyze them using convolutional layers to extract features related to movement behaviors. The primary metric will be the percentage change (minutes a day) in real-world daytime movement of the trained limb, computed relative to baseline.

This outcome reflects changes in functional mobility resulting from the motor training intervention.

Attendance will be measured as the number of times participants log into the tablet app during the intervention.

Unit of Measure: Number of logins

Adherence will be assessed by the average length of each daily training session. This outcome reflects how long participants remain engaged with the motor training program once they log in.

Unit of Measure: Minutes

This measure reflects adherence by quantifying the active movement time of the trained extremity during each training session.

Unit of Measure: Minutes

For participants in the Music Motion Feedback group, adherence will also be evaluated by measuring the time spent at or above a predefined movement intensity threshold.

Unit of Measure: Minutes

In participants engaged in upper extremity training, the ARAT will be used to evaluate upper extremity functional capacity. It consists of 19 items assessing grasp, grip, pinch, and gross movement. Scores range from 0 to 57, with higher scores indicating better arm function.

Time Frame: Baseline, 12 weeks (end of intervention), and 24 weeks (follow-up) Unit of Measure: ARAT total score (0-57)

Participants engaged in upper extremity training will have the isometric strength of the targeted upper extremity measured using a handheld dynamometer. They will perform maximum voluntary contractions against resistance, and the peak force will be recorded.

Unit of Measure: Kilograms of force (kgf)

Participants who are engaged in upper extremity training will have their grip strength of the targeted upper limb measured using a hand dynamometer. The highest of three attempts will be recorded.

Unit of Measure: Kilograms of force (kgf)

In participants who are engaged in lower extremity training the GMFM-66 will be used to assess gross motor function. It evaluates activities such as standing, walking, and running. Scores range from 0 to 100, with higher scores indicating better function.

Unit of Measure: GMFM-66 total score (0-100)

The isometric strength of the trained leg will be assessed using a handheld dynamometer. Participants will perform maximum voluntary contractions, and peak force will be recorded.

Unit of Measure: Kilograms of force (kgf)

The 6MWT will be used to evaluate walking endurance and aerobic capacity. Participants engaged in lower extremity training will walk as far as possible in six minutes along a flat, indoor course.

Time Frame: Baseline, 12 weeks, and 24 weeks Unit of Measure: Meters walked

During a semi-structured interview, participants identify an activity they want, need, or are expected to perform and set one long-term goal. They then rate their performance of that activity on a 0-10 scale, where 0 = not able to perform the activity at all and 10 = able to perform it extremely well.

Unit of Measure: Performance score (0-10 scale)

In the same semi-structured interview, participants rate their satisfaction with their performance of the selected activity on a 0-10 scale, where 0 = not satisfied at all and 10 = extremely satisfied.

Unit of Measure: Satisfaction score (0-10 scale)

Time spent walking will be measured using 3-axis accelerometer and gyroscope data collected from seven IMUs worn continuously for 72 hours. Data will be analyzed using a validated deep-learning neural network (Novosel et al., 2023) to classify walking behavior.

Unit of Measure: Minutes per day

Standing time will be extracted from IMU data collected during 72 hours of wear. Signals will be analyzed via a validated deep-learning neural network (Novosel et al., 2023) Time Frame: Baseline, 12 weeks, and 24 weeks Unit of Measure: Minutes per day

Sitting time will be recorded using IMUs worn on the sternum, wrists, thighs, and lower legs after 72 hours of wear and analysed with a validated deep-learning neural network (Novosel et al., 2023)

Unit of Measure: Minutes per day

Time spent lying will be measured using 3-axis accelerometer and gyroscope data collected from seven IMUs worn continuously for 72 hours. Data will be analyzed using a validated deep-learning neural network (Novosel et al., 2023) to classify walking behavior.

Unit of Measure: Minutes per day

Posture transitions (e.g., sit-to-stand) will be identified from 72-hour IMU data using a validated deep-learning neural network (Novosel et al., 2023)

Unit of Measure: Number of transitions per day

The movement ratio between the trained and non-trained extremity will be calculated from IMU signals using using a validated deep-learning neural network (Novosel et al., 2023)

Unit of Measure: Ratio (trained / non-trained limb)

Adverse events include both serious (e.g. death, hospitalization) and non-serious adverse events (e.g. pain, fatigue).

Unit of Measure: Number of events