During the COVID-19 pandemic, significant challenges arose in the on-site monitoring of patients’ clinical signs due to healthcare system overload, lack of resources, and the need for social distancing. These obstacles hindered readiness in identifying and responding promptly to cases, highlighting the importance of investments in healthcare infrastructure and technologies for effective monitoring in emergency situations. This study explores the use of a wearable, wireless, and scalable system for remote monitoring of physiotherapy sessions with an emphasis on applying human activity recognition. It employs a variety of sensors and equipment for the classification of physiotherapeutic exercises from a distance. The sensors and equipment provide data to a web platform that allows, for example, determining posture and classifying the activity performed by the patient through measuring the angulation between body limbs. This platform includes the design of wearable accessories, 3D-printed, portable, and wireless hardware construction. The web part consists of a remote server, a microservices environment including the provision of a web portal ( https://bionet.ufpe.br ) for user interaction, as well as data storage and processing, providing the information. Currently, a testing protocol is under development to be executed by volunteer physiotherapy specialists and their respective patients, with approval from the ethics committee (CAAE 71106023.0.0000.5208).