Abstract
OBJECTIVE: To evaluate the efficacy of a clinical data-driven intelligent diagnosis and treatment model in chronic obstructive pulmonary disease (COPD) home rehabilitation, focusing on improving patient outcomes through personalized interventions. METHODS: A total of 82 COPD patients (control group) received conventional care from January to September 2023. An additional 86 patients (intelligent group) were managed via a cloud platform integrating wearable devices (smart bracelets, pulmonary function meters) and symptom tracking apps from October 2023 to July 2024. Real-time data (blood oxygen, heart rate, FEV1%, FVC) were uploaded daily, with long short-term memory -based acute exacerbation alerts. Personalized rehabilitation, medication, and nutrition guidance were delivered via mobile platforms, alongside biweekly video consultations. A 6-month follow-up assessed pulmonary function (FVC, FEV1%, MMEF, PEF), cardiac function (LVEDD, LVEF), 6MWD, dyspnea (mMRC), quality of life (SGRQ), compliance, and exacerbation events. RESULTS: At 3-6 months, the intelligent group showed significant improvements in pulmonary function (higher FVC, FEV1%, MMEF, PEF; all P < 0.05), cardiac function (improved LVEDD, LVEF, 6MWD; all P < 0.05), symptoms (lower mMRC, SGRQ scores; both P < 0.05), and compliance (higher data upload and intervention response rates; both P < 0.001). Exacerbations were reduced, with fewer events and shorter hospitalizations (all P < 0.05). CONCLUSIONS: The intelligent model improves COPD rehabilitation through real-time monitoring and personalized care, significantly enhancing pulmonary and cardiac function, exercise capacity, and quality of life, while reducing exacerbations and hospitalizations. This model holds potential for broader clinical adoption.