AI Chatbot-Assisted Aerobic Exercise Guided by the Transtheoretical Model and Self-determination Theory With Machine Learning Algorithms to Provide Real-time, Personalized Feedback in Traumatic Brain Injury

NCT07783828 | Recruiting

• 1 other identifier: AI chatbot-assisted exercise
• Study Type: interventional
• Target: 50
• Locations: 1 country
• Sites: 1

Brief Summary

This study aims to develop and evaluate an AI-driven Personalized Exercise Feedback Program (AI-PEF) to enhance health outcomes in mTBI patients. Methods: AI-PEF integrates the transtheoretical model and self-determination theory with machine learning algorithms to provide real-time, personalized feedback. A phased randomized controlled trial will be conducted: Phase I,small-Scale Pilot Study (Early Feedback): A single-arm pilot study with 5 patients will be conducted over four weeks to gather feedback on usability, engagement, and content clarity.Phase II assesses the program's impact on functional capacity, sleep quality and depressive symptoms, with 50 participants in a 2-arm randomized trial in 3 months.Phase III will assess the program's impact on functional capacity, sleep quality and depressive symptoms, with 50 participants of waitlist control evaluation in a two-arm randomized trial.

Conditions

Tramatic Brain Injury

Outcome Measures

Primary Outcomes (20)

• 6-Minute Walk Test

  Functional capacity will be evaluated using the 6-Minute Walk Test (6MWT), which measures the distance walked in six minutes as an indicator of functional exercise capacity. A longer walking distance(meter) indicates better functional capacity.

  T0-baseline

• 6-Minute Walk Test

  Functional capacity will be evaluated using the 6-Minute Walk Test (6MWT), which measures the distance walked in six minutes as an indicator of functional exercise capacity. A longer walking distance(meter) indicates better functional capacity.

  T1(1 month later)

• 6-Minute Walk Test

  Functional capacity will be evaluated using the 6-Minute Walk Test (6MWT), which measures the distance walked in six minutes as an indicator of functional exercise capacity. A longer walking distance(meter) indicates better functional capacity.

  T2(2 month later)

• 6-Minute Walk Test

  Functional capacity will be evaluated using the 6-Minute Walk Test (6MWT), which measures the distance walked in six minutes as an indicator of functional exercise capacity. A longer walking distance(meter) indicates better functional capacity.

  T3(3 month later)

• 6-Minute Walk Test

  Functional capacity will be evaluated using the 6-Minute Walk Test (6MWT), which measures the distance walked in six minutes as an indicator of functional exercise capacity. A longer walking distance(meter) indicates better functional capacity.

  T4(6 month later)

• Adherence

  Exercise adherence will be defined as the percentage of prescribed sessions completed during the 12-week intervention, based on three 30-minute sessions per week (36 sessions in total). The Garmin Vivosmart 5 will record exercise frequency, duration, and intensity. A session will be considered completed when the participant exercises for at least 30 minutes and reaches Garmin heart rate zone 2 or higher. Adherence will be calculated by summing completed sessions, with a maximum of three sessions counted per week, dividing by 36, and multiplying by 100. Scores range from 0% to 100%, with higher percentages indicating greater adherence. Borg Rating of Perceived Exertion (RPE) will complement intensity assessment, particularly for older participants or those with heart rate or blood pressure concerns. An RPE of ≥12 indicates moderate intensity, while ≥11 may be encouraged initially, followed by gradual progression as tolerated.

  T0-baseline

• Adherence

  Exercise adherence will be defined as the percentage of prescribed sessions completed during the 12-week intervention, based on three 30-minute sessions per week (36 sessions in total). The Garmin Vivosmart 5 will record exercise frequency, duration, and intensity. A session will be considered completed when the participant exercises for at least 30 minutes and reaches Garmin heart rate zone 2 or higher. Adherence will be calculated by summing completed sessions, with a maximum of three sessions counted per week, dividing by 36, and multiplying by 100. Scores range from 0% to 100%, with higher percentages indicating greater adherence. Borg Rating of Perceived Exertion (RPE) will complement intensity assessment, particularly for older participants or those with heart rate or blood pressure concerns. An RPE of ≥12 indicates moderate intensity, while ≥11 may be encouraged initially, followed by gradual progression as tolerated.

  T1(1 month later)

• Adherence

  Exercise adherence will be defined as the percentage of prescribed sessions completed during the 12-week intervention, based on three 30-minute sessions per week (36 sessions in total). The Garmin Vivosmart 5 will record exercise frequency, duration, and intensity. A session will be considered completed when the participant exercises for at least 30 minutes and reaches Garmin heart rate zone 2 or higher. Adherence will be calculated by summing completed sessions, with a maximum of three sessions counted per week, dividing by 36, and multiplying by 100. Scores range from 0% to 100%, with higher percentages indicating greater adherence. Borg Rating of Perceived Exertion (RPE) will complement intensity assessment, particularly for older participants or those with heart rate or blood pressure concerns. An RPE of ≥12 indicates moderate intensity, while ≥11 may be encouraged initially, followed by gradual progression as tolerated.

  T2(2 month later)

• Adherence

  Exercise adherence will be defined as the percentage of prescribed sessions completed during the 12-week intervention, based on three 30-minute sessions per week (36 sessions in total). The Garmin Vivosmart 5 will record exercise frequency, duration, and intensity. A session will be considered completed when the participant exercises for at least 30 minutes and reaches Garmin heart rate zone 2 or higher. Adherence will be calculated by summing completed sessions, with a maximum of three sessions counted per week, dividing by 36, and multiplying by 100. Scores range from 0% to 100%, with higher percentages indicating greater adherence. Borg Rating of Perceived Exertion (RPE) will complement intensity assessment, particularly for older participants or those with heart rate or blood pressure concerns. An RPE of ≥12 indicates moderate intensity, while ≥11 may be encouraged initially, followed by gradual progression as tolerated.

  T3(3 month later)

• Adherence

  Exercise adherence will be defined as the percentage of prescribed sessions completed during the 12-week intervention, based on three 30-minute sessions per week (36 sessions in total). The Garmin Vivosmart 5 will record exercise frequency, duration, and intensity. A session will be considered completed when the participant exercises for at least 30 minutes and reaches Garmin heart rate zone 2 or higher. Adherence will be calculated by summing completed sessions, with a maximum of three sessions counted per week, dividing by 36, and multiplying by 100. Scores range from 0% to 100%, with higher percentages indicating greater adherence. Borg Rating of Perceived Exertion (RPE) will complement intensity assessment, particularly for older participants or those with heart rate or blood pressure concerns. An RPE of ≥12 indicates moderate intensity, while ≥11 may be encouraged initially, followed by gradual progression as tolerated.

  T4(6 month later)

• The Chinese Version of the Sport Motivation Scale-II

  The Chinese version of the Sport Motivation Scale-II (CSMS-II) will be used to assess six types of motivation: intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation. The scale consists of 18 items rated on a 7-point Likert scale from 1 to 7. The subscales include intrinsic motivation (items 3, 9, and 17), integrated regulation (items 4, 11, and 14), identified regulation (items 6, 12, and 18), introjected regulation (items 1, 7, and 16), external regulation (items 5, 8, and 15), and amotivation (items 2, 10, and 13). Each subscale is scored separately by calculating the mean of its three items. Each subscale score ranges from 1 to 7, with higher scores indicating a stronger level of the corresponding motivational regulation.

  T0-baseline

• The Chinese Version of the Sport Motivation Scale-II

  The Chinese version of the Sport Motivation Scale-II (CSMS-II) will be used to assess six types of motivation: intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation. The scale consists of 18 items rated on a 7-point Likert scale from 1 to 7. The subscales include intrinsic motivation (items 3, 9, and 17), integrated regulation (items 4, 11, and 14), identified regulation (items 6, 12, and 18), introjected regulation (items 1, 7, and 16), external regulation (items 5, 8, and 15), and amotivation (items 2, 10, and 13). Each subscale is scored separately by calculating the mean of its three items. Each subscale score ranges from 1 to 7, with higher scores indicating a stronger level of the corresponding motivational regulation.

  T1(1 month later)

• The Chinese Version of the Sport Motivation Scale-II

  The Chinese version of the Sport Motivation Scale-II (CSMS-II) will be used to assess six types of motivation: intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation. The scale consists of 18 items rated on a 7-point Likert scale from 1 to 7. The subscales include intrinsic motivation (items 3, 9, and 17), integrated regulation (items 4, 11, and 14), identified regulation (items 6, 12, and 18), introjected regulation (items 1, 7, and 16), external regulation (items 5, 8, and 15), and amotivation (items 2, 10, and 13). Each subscale is scored separately by calculating the mean of its three items. Each subscale score ranges from 1 to 7, with higher scores indicating a stronger level of the corresponding motivational regulation.

  T2(2 month later)

• The Chinese Version of the Sport Motivation Scale-II

  The Chinese version of the Sport Motivation Scale-II (CSMS-II) will be used to assess six types of motivation: intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation. The scale consists of 18 items rated on a 7-point Likert scale from 1 to 7. The subscales include intrinsic motivation (items 3, 9, and 17), integrated regulation (items 4, 11, and 14), identified regulation (items 6, 12, and 18), introjected regulation (items 1, 7, and 16), external regulation (items 5, 8, and 15), and amotivation (items 2, 10, and 13). Each subscale is scored separately by calculating the mean of its three items. Each subscale score ranges from 1 to 7, with higher scores indicating a stronger level of the corresponding motivational regulation.

  T3(3 month later)

• The Chinese Version of the Sport Motivation Scale-II

  The Chinese version of the Sport Motivation Scale-II (CSMS-II) will be used to assess six types of motivation: intrinsic motivation, integrated regulation, identified regulation, introjected regulation, external regulation, and amotivation. The scale consists of 18 items rated on a 7-point Likert scale from 1 to 7. The subscales include intrinsic motivation (items 3, 9, and 17), integrated regulation (items 4, 11, and 14), identified regulation (items 6, 12, and 18), introjected regulation (items 1, 7, and 16), external regulation (items 5, 8, and 15), and amotivation (items 2, 10, and 13). Each subscale is scored separately by calculating the mean of its three items. Each subscale score ranges from 1 to 7, with higher scores indicating a stronger level of the corresponding motivational regulation.

  T4(6 month later)

• The mHealth App usability questionnaire

  The Mobile App Usability Questionnaire (MAUQ) assesses usability across three dimensions: usability and satisfaction, system information arrangement, and system efficiency. A Chinese version of the questionnaire was developed in 2022, with an internal consistency reliability (Cronbach's α) of 0.988 and an internal consistency range of 0.845-0.931. The test-retest reliability ranges from 0.828 to 0.918. A total of 21 items, Likert scale scores ranged from 1 (strongly agree) to 7 (strongly disagree), indicating the lower scores the better usability.

  T0-baseline

• The mHealth App usability questionnaire

  The Mobile App Usability Questionnaire (MAUQ) assesses usability across three dimensions: usability and satisfaction, system information arrangement, and system efficiency. A Chinese version of the questionnaire was developed in 2022, with an internal consistency reliability (Cronbach's α) of 0.988 and an internal consistency range of 0.845-0.931. The test-retest reliability ranges from 0.828 to 0.918. A total of 21 items, Likert scale scores ranged from 1 (strongly agree) to 7 (strongly disagree), indicating the lower scores the better usability.

  T1(1 month later)

• The mHealth App usability questionnaire

  The Mobile App Usability Questionnaire (MAUQ) assesses usability across three dimensions: usability and satisfaction, system information arrangement, and system efficiency. A Chinese version of the questionnaire was developed in 2022, with an internal consistency reliability (Cronbach's α) of 0.988 and an internal consistency range of 0.845-0.931. The test-retest reliability ranges from 0.828 to 0.918. A total of 21 items, Likert scale scores ranged from 1 (strongly agree) to 7 (strongly disagree), indicating the lower scores the better usability.

  T2(2 month later)

• The mHealth App usability questionnaire

  The Mobile App Usability Questionnaire (MAUQ) assesses usability across three dimensions: usability and satisfaction, system information arrangement, and system efficiency. A Chinese version of the questionnaire was developed in 2022, with an internal consistency reliability (Cronbach's α) of 0.988 and an internal consistency range of 0.845-0.931. The test-retest reliability ranges from 0.828 to 0.918. A total of 21 items, Likert scale scores ranged from 1 (strongly agree) to 7 (strongly disagree), indicating the lower scores the better usability.

  T3(3 month later)

• The mHealth App usability questionnaire

  The Mobile App Usability Questionnaire (MAUQ) assesses usability across three dimensions: usability and satisfaction, system information arrangement, and system efficiency. A Chinese version of the questionnaire was developed in 2022, with an internal consistency reliability (Cronbach's α) of 0.988 and an internal consistency range of 0.845-0.931. The test-retest reliability ranges from 0.828 to 0.918. A total of 21 items, Likert scale scores ranged from 1 (strongly agree) to 7 (strongly disagree), indicating the lower scores the better usability.

  T4(6 month later)

Secondary Outcomes (50)

• The Pittsburgh Sleep Quality Index

  T0-baseline

• The Beck Depression Inventory-II

  T0-baseline

• The Rivermead Post-Concussion Symptoms Questionnaire

  T0(baseline)

• The Godin-Shephard Leisure Time Physical Activity Questionnaire

  T0( baseline)

• The World Health Organization Quality of Life Scale-brief

  T0( baseline)

Study Arms (2)

AI-PEF

EXPERIMENTAL

the AI-PEF group, receiving a machine learning-powered personalized exercise program.

Behavioral: AI-PEF

Active control group

ACTIVE COMPARATOR

the Active control group, receiving general exercise recommendations as part of standard care in a 1:1 ratio.

Behavioral: Active Control

Interventions

AI-PEF BEHAVIORAL

Participants will follow a 6months walking program with five 30-minute sessions per week, progressively increasing intensity based on heart rate and effort indices. Garmin fitness trackers will monitor adherence and intensity. Remote guidance will support participants via weekly digital health messages on the LINE app, offering personalized feedback, goal reinforcement, and lifestyle recommendations. Participants will also receiThese AI feedback implementations will be tailored to complement the in-person education and will include reminders of the individualized goals set during the remote exercise intervention and consultation . The digital component will also offer a platform for patients to share their progress and seek further guidance during the scheduled in-person consultation. The use of digital AI feedback aligns with the trend of integrating technology into healthcare services, providing a convenient and accessible modality for supporting parents engaging in regular exercise.

AI-PEF

Active Control BEHAVIORAL

Participants in the active control group will receive standard TBI care, including Garmin-based self-monitoring and routine clinic visits at baseline (T0), 1 months (T1), 2months(T2),and 3 months (T3). These visits will include general health education and recommendations on daily physical activity. No personalized or digital exercise strategies will be provided. Participants completing the 3-month protocol will have the option to access the AI-PEF program. When active control group participants completing the 3-month protocol will have the option to access the AI-PEF program.Then participants will use AI-PEF program to be the waitlist control group after 3 months to 6months, with assessments conducted at 6 months (T4).

Active control group

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Eligible participants are patients aged over 18 with mild TBI (GCS 13-15)
• who can walk independently,
• reside in the Greater Taipei area,
• and possess sufficient Chinese or Taiwanese language proficiency to understand the trial
• complete self-administered questionnaires.

You may not qualify if:

• include individuals with severe medical conditions (e.g., respiratory failure, epilepsy, psychiatric disorders), musculoskeletal or neurological impairments
• hindering physical activity in the 6-minute walk test,
• cognitive impairments (MMSE \< 24),
• frontal lobe injuries or penetrating injury causing significant psychological dysfunction.
• Patients regularly engaging in moderate-to-high-intensity aerobic exercise or participating in other studies will also be excluded to avoid bias.

Sponsors & Collaborators

• National Defense Medical Center, Taiwan: lead

Study Sites (1)

Tri-Service General Hospital

Taipei, 11490, Taiwan

RECRUITING

Central Study Contacts

Hui Hsun Chiang, Professor

CONTACT

886-2-87923100; sheisvivian@gmail.com

Hui Hsun Chiang, Professor

CONTACT

886-921060383; huihsunchiang@mail.ndmctsgh.edu.tw

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: TRIPLE
• Who Masked: PARTICIPANT, CARE PROVIDER, OUTCOMES ASSESSOR
• Purpose: SUPPORTIVE CARE
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor

Model Details: Phase 1: A single-arm pilot study with 5 patients will be conducted over four weeks to gather feedback on usability, engagement, and content clarity. Phase 2: Evaluate the health outcomes, including functional capacity, sleep quality and depressive symptoms of the AI-PEF intervention in improving mTBI patients. Phase 3:Examine the impact of the AI-PEF intervention on health outcomes through a waitlist 2-arm RCT.

Study Record Dates

• First Submitted: May 20, 2025
• First Posted: August 25, 2026
• Study Start: August 14, 2025
• Primary Completion (Estimated): December 31, 2030
• Study Completion (Estimated): August 31, 2031
• Last Updated: August 25, 2026

Record last verified: 2026-08

Data Sharing

• IPD Sharing: Will not share

Locations

TW (1)