Healthcare Recommender System Based on Medical Specialties, Patient Profiles, and Geospatial Information

The global outburst of COVID-19 introduced severe issues concerning the capacity and adoption of healthcare systems and how vulnerable citizen classes might be affected. The pandemic generated the most remarkable transformation of health services, appropriating the increase in new information and communication technologies to bring sustainability to health services. This paper proposes a novel, methodological, and collaborative approach based on patient-centered technology, which consists of a recommender system architecture to assist the health service level according to medical specialties. The system provides recommendations according to the user profile of the citizens and a ranked list of medical facilities. Thus, we propose a health attention factor to semantically compute the similarity between medical specialties and offer medical centers with response capacity, health service type, and close user geographic location. Thus, considering the challenges described in the state-of-the-art, this approach tackles issues related to recommenders in mobile devices and the diversity of items in the healthcare domain, incorporating semantic and geospatial processing. The recommender system was tested in diverse districts of Mexico City, and the spatial visualization of the medical facilities filtering by the recommendations is displayed in a Web-GIS application.

Automated summary

This paper proposes a patient-centered recommender system architecture to address the capacity and vulnerability issues in healthcare systems due to COVID-19. The system provides recommendations based on user profiles and a ranked list of medical facilities, incorporating semantic and geospatial processing. The approach was tested in diverse districts of Mexico City, and the results were visualized in a Web-GIS application.