Artificial Intelligence (AI)-Based Personalized Exercise Coaching for Patients With Breast Cancer

Participants assigned to this group will use an AI-based exercise coaching system for 6 months. The system will generate individualized exercise recommendations based on participant characteristics and health-related information. Exercise data will be collected and uploaded to the system, allowing exercise recommendations to be dynamically adjusted according to participants' exercise performance and status. Participants will also receive regular follow-up during the intervention period.

Participants will receive individualized exercise recommendations generated by an AI-based exercise coaching system. The system provides personalized exercise guidance and uses exercise data collected to monitor exercise performance and dynamically adjust recommendations. The intervention will be delivered for 6 months.

Participants assigned to the control group will receive conventional exercise guidance from healthcare professionals based on routine clinical practice and professional experience, including general recommendations such as approximately 150 minutes of moderate-intensity aerobic exercise per week. Participants will receive regular follow-up during the 6-month study period.

General exercise recommendations will be provided by healthcare professionals according to routine clinical practice.

Exercise adherence will be calculated as the ratio of the participant's actual exercise frequency to the recommended exercise frequency during the intervention period, expressed as a percentage. Exercise adherence will be compared between the AI-based exercise coaching group and the conventional exercise guidance group.

Health-related quality of life will be assessed using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30). Scores will be calculated according to the EORTC scoring procedures, with domain scores transformed to a 0-100 scale. Higher scores on functional scales and global health status indicate better functioning or quality of life, whereas higher scores on symptom scales indicate greater symptom burden.

Anxiety symptoms will be assessed using the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS-A). The HADS-A score ranges from 0 to 21. Higher scores indicate more severe anxiety symptoms.

Depressive symptoms will be assessed using the depression subscale of the Hospital Anxiety and Depression Scale (HADS-D). The HADS-D score ranges from 0 to 21. Higher scores indicate more severe depressive symptoms.

Peak oxygen uptake (VO₂peak) will be measured by cardiopulmonary exercise testing (CPET) on an electronically braked cycle ergometer, using a CORTEX MetaLyzer 3B-R3 metabolic cart, during a symptom-limited incremental exercise protocol. VO₂peak is recorded at the highest oxygen uptake achieved during the test. The unit of measurement is mL/kg/min.

Muscle strength of upper- and lower-extremity groups will be assessed bilaterally by manual muscle testing (MMT) using the Medical Research Council (MRC) 0-5 grading scale. The unit of measurement is MRC muscle grade (0-5). Higher grades indicate greater muscle strength.

Active and passive joint range of motion (AROM and PROM) of the shoulder, elbow and wrist will be measured using a universal goniometer, on both left and right sides. The unit of measurement is degrees (°).

Low-density lipoprotein cholesterol will be measured from fasting venous blood samples using an automatic biochemical analyzer. The unit of measurement is mmol/L.

Total cholesterol will be measured from fasting venous blood samples using an automatic biochemical analyzer. The unit of measurement is mmol/L.

Fasting C-peptide will be measured from venous blood samples collected after at least 8 hours of overnight fasting, using a chemiluminescent immunoassay analyzer. The unit of measurement is ng/mL.

The number and proportion (%) of participants experiencing exercise-related adverse events, including muscle strain, joint sprain, falls, and other exercise-related injuries, will be recorded during the intervention period. The outcome is reported as the percentage of participants with at least one exercise-related adverse event. The unit of measurement is percentage (%).