| DOI | Trouver le DOI : https://doi.org/10.1109/ICHI52183.2021.00038 |
|---|
| Auteur | Rechercher : Rivest-Hénault, David1; Rechercher : Pagiatakis, Catherine1; Rechercher : Bernhardt, Richard1; Rechercher : Vaughan, Thomas1; Rechercher : Falardeau, Bruno1; Rechercher : Smith, Michael S. D.1; Rechercher : Jiang, Di1 |
|---|
| Affiliation | - Conseil national de recherches du Canada. Dispositifs médicaux
|
|---|
| Format | Texte, Article |
|---|
| Conférence | 2021 IEEE 9th International Conference on HealthCare Informatics (ICHI), August 9-12, 2021, Victoria, BC, Canada |
|---|
| Sujet | contactless sensing; PPG; rPPG; RGB camera; remote patient monitoring; heart rate |
|---|
| Résumé | Remote patient monitoring has only become more relevant in the current pandemic context. Contactless sensing extracts vital signs (ex. heart rate) and other physiological information from signals obtained from non-contact hardware such as video cameras. Such a technology could have a significant impact on the healthcare system by enabling a physician to remotely acquire quantitative measurements of a patient. This could help prevent sudden deterioration of patient condition, improve the efficiency and support for medical decision making and diagnostics, and reduce costs and hospitalizations. In this work, a practical and fully automatic video-based contactless monitoring framework for the evaluation of patient heart rate is presented. The framework allows the calculation of heart rate in quasi realtime and with a high accuracy, using a consumer grade laptop equipped with a usb-connected webcam. The proposed heart rate assessment algorithm combines state-of-the art methods in video- and signal-processing for contactless sensing; to attain high accuracy predictions, an additional probabilistic formulation that improves the temporal consistency is proposed and used along with live error rejection. To validate the framework, twentytwo subjects were recorded under varying physiological states (total of 111 two-minute long videos). Heart rate predictions were compared to medical-grade sensors and showed an accuracy of -0.17±1.81 bpm with a MAE of 0.91 bpm. |
|---|
| Date de publication | 2021-08-09 |
|---|
| Maison d’édition | IEEE |
|---|
| Dans | |
|---|
| Langue | anglais |
|---|
| Publications évaluées par des pairs | Oui |
|---|
| Numéro du CNRC | NRC-MD-2021-004 |
|---|
| Exporter la notice | Exporter en format RIS |
|---|
| Signaler une correction | Signaler une correction (s'ouvre dans un nouvel onglet) |
|---|
| Identificateur de l’enregistrement | 87a9660d-1b51-4760-bad0-5afee3606302 |
|---|
| Enregistrement créé | 2021-10-04 |
|---|
| Enregistrement modifié | 2021-11-29 |
|---|
