Resumen
Datos solicitados: Grabación simultánea de registro ECG, pulsometría y vídeo de un dedo del paciente con un móvil
Objetivo del proyecto: Desarrollo de apps para el seguimiento de pacientes tras el alta
Requester
Name: Yaël Frégier
Institution: Université d’Artois, Lens, France
Web: Research web
Description of project:
Develop an app that remotely measures ECG/BVP signals and incorporate it to open source platforms as Covidom and Covid [2]. Our deep-learning model predicts the BVP from a video of a subject’s face (Figure 1). We want to adapt it with the requested data to use the video of a finger and test if the model can be modified to predict some ECG features ([1] suggests that it is feasible).
Figure 1: Comparison between blood volume pulse measured by a dedicated device (orange) and reconstructed from video (blue).
Terms of collaboration:
The requester would like to establish a collaboration including the collection of the data, as well as advising on the medical side of the project (such as suggestion of use cases that they haven’t yet foreseen). This collaboration may include co-authorship in any eventual publication.
Detailed description of data:
We would need the measurements of the ECG and BVP of a large number of patients and the recordings of their finger placed on the camera of a mobile device. Ideally the camera would record at 30 frames per second and the measurements of ECG and BVP would have to be made at the same moment as the frames, but we can also interpolate the ECG/BVP data at a later moment if this setting is not possible. Video recording and the ECG/BVP measurements would have to start at the same moment, or at least the lag would have to be known, in order to match successfully the BVP/ECG signals to the videos.
Handling of personal/confidential information:
All data can be anonymized before sending to the requester (fingerprints cannot be extracted from this type of video recordings, so it is not possible to identify a subject form this data). The requester would be willing to either have the data publicly available (open source) or to share ownership.
Status of ethical approval:
The requester will work with anonymized datasets. They are willing to work with the data provider to obtain any necessary ethical approvals to collect the data.
References:
[1] N. Mahri, K. B. Gan, and M. A. M. Ali. Extracting features similar to qt interval from second derivatives
of photoplethysmography: A feasibility study. In 2014 IEEE Conference on Biomedical Engineering and
Sciences (IECBES), pages 470{473, 2014.
[2] P. Ravaud, F. le Ouay, E. Depaulis, A. Huckert, B. Vegreville, and V.-T. Tran. Reconfiguring health
services to reduce the workload of caregivers during the covid-19 outbreak using an open-source scalable
platform for remote digital monitoring and coordination of care in hospital command centres.
Nota: En caso necesario, Crowdfight COVID-19 puede proporcionar ayuda para el procesamiento, preparación o envío de los datos. Esto incluye asesoramiento técnico, apoyo de expertos en bases de datos, o voluntarios que realicen tareas manuales (organización de datos, etc.).