Monitoring and detection of toothbrushing with smart watches and artificial intelligence.
Loading...
Files
Description: Resumen de la ponencia
Identifiers
Publication date
Reading date
Collaborators
Advisors
Tutors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Share
Department/Institute
Abstract
This work proposes to combine wearables and the use of artificial intelligence techniques to detect manual mobility patterns caused by brushing teeth. Specifically, the article describes and evaluate a system based on convolutional neural networks, able to identify brushing gestures from samples of few seconds of inertial accelerometry signals gathered by wrist-worn devices. The architecture is systematically trained and validated with the signals provided by different public databases which were collected when different experimental subjects executed different manual actions. The results show the effectiveness of the detector, since it reaches a sensitivity and specificity greater than 95% when applied to discriminate brushing from other hand actions. In addition, the system is re-trained and assessed with the real-life samples captured by a smartwatch, where the neural model could be implemented to operate and produce real-time decisions.
Description
It has been estimated that oral and dental diseases affect almost half of humanity, largely due to the infrequency with which a considerable proportion of the population brush their teeth (less than twice a day, with a regularity that markedly decreases for older ages). In this context, the automatic monitoring of dental hygiene routines may be of great interest, since it can help promote healthy habits and remind the user (especially older persons or patients in the initial stages of dementia) of the need to brushing her/his teeth after each meal.In this sense, although initially envisaged to track sporting performance, current smartwatches and smartbands have found a relevant field in HAR (Human Activity Recognition) systems, especially in those applications intended to supervise health status and personal well-being. Thus, these wearables offer a low-cost, non-invasive tool, which is already fully integrated into our daily lives, capable of informing us at all times about the evolution of diverse biosignals or health parameters and even generating alerts in case a medical alarm is suspected. This work proposes to combine wearables and the use of artificial intelligence techniques to detect manual mobility patterns caused by brushing teeth. Specifically, the article describes and evaluate a system based on convolutional neural networks, able to identify brushing gestures from samples of few seconds of inertial accelerometry signals gathered by wrist-worn devices. The architecture is systematically trained and validated with the signals provided by different public databases which were collected when different experimental subjects executed different manual actions. The results show the effectiveness of the detector, since it reaches a sensitivity and specificity greater than 95% when applied to discriminate brushing from other hand movements. In addition, the system is re-trained and assessed with the real-life samples captured by a smartwatch, where the neural model is implemented to operate and produce real-time decisions