Monitoring of frequent exacerbators with COPD through the application of a virtual assistant with artificial intelligence

Le résumé fourni par la source

Machine learning with algorithms applications can facilitate remote patient monitoring. Our objective was to evaluate the implementation of a virtual case manager in the follow-up of exacerbator COPD Prospective observational pilot study. COPD patients with at least 2 outpatient exacerbations or 1 hospital admission in the previous year were included. All patients received a weekly call from the virtual manager answering questions about their health status and symptoms. These responses could generate alerts of different severity that were reviewed by the research team, leading to medical intervention if required To date, 104 patients have been included with a mean follow-up of 4.1(SD 2.4) months. 65% were men with a mean age of 66(SD 12.2) years. The mean FEV1(%) was 38.9% (SD 13.5). Between 01/3/2024 - 31/01/2025, the virtual manager made 2,304 calls with an average duration of 5:57minuts, an adherence rate of 96% and a generation of 2904 alerts(32% severe). Those alerts generated 1158 medical interventions(n): review medical history(597), telephone call(367), medication adjustment(40), prescription of corticosteroids(25) and antibiotics(28), outpatient visit(27) and emergency visit(12). The most frequent alerts were(n): increased need for rescue medication(552) and changes in the colour of sputum(444). However, the alerts that generate most medical interventions were general malaise(72%), clinical observations(88%) and emergency visit(81%). 58 exacerbations were detected early The implementation of a virtual case manager in the follow-up of COPD exacerbator patients has good adherence, improves follow-up and early detection of exacerbations.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.