Human Activity Recognition (HAR) plays an important role in the automation of various tasks related to activity tracking in such areas as healthcare and eldercare (telerehabilitation, telemonitoring), security, ergonomics, entertainment (fitness, sports promotion, human–computer interaction, video games), and intelligent environments. This paper tackles the problem of real-time recognition and repetition counting of 12 types of exercises performed during athletic workouts. Our approach is based on the deep neural network model fed by the signal from a 9-axis motion sensor (IMU) placed on the chest. The model can be run on mobile platforms (iOS, Android). We discuss design requirements for the system and their impact on data collection protocols. We present architecture based on an encoder pretrained with contrastive learning. Compared to end-to-end training, the presented approach significantly improves the developed model’s quality in terms of accuracy (F1 score, MAPE) and robustness (false-positive rate) during background activity. We make the AIDLAB-HAR dataset publicly available to encourage further research.
Aim: The aim of this paper is to present the results of the validation of AIDMED as a telemedical system, i.e. its capability in faithful registration of biomedical signals, its acquisition in a telemedical scenario and its representation in online application. Usability of sucha tool for a dedicated population was also assessed.Methods: We describe and discuss functionalities provided by AIDMED. We perform a series of experiments where we measure biological signals with AIDMED and with a reference device. We provide statistical analysis of experiments. We also compare the functionality of AIDMED with other similar solutions. We discuss the usability of AIDMED in tele observation of COVID-19 patients.Results: We show diagnostic equivalence of AIDMED device and reference devices.Moreover, we indicate advantages of AIDMED system (as task management and patient’s feedback via mobile app) for at home telemonitoring in comparison to standard of care.Conclusions: AIDMED system provides an integrated platform which enables observation of COVID-19, cardiological and pulmonary patients and many more. Thus, an opportunity for both better quality of care and better subjective patient satisfaction with use of AIDMED has got a solid foundation.