Atrial Fibrillation
Conditions
Brief summary
The project focuses on the evaluation of a novel, contactless monitoring technology to measure the blood pulsatile signal based on the video recording of an individual's face. The variability of the pulse rate is computed to identify the presence of atrial fibrillation (AF). We propose to enroll 315 patients with symptomatic AF, paroxysmal or persistent, who go through successful radiofrequency ablation or electrical cardioversion. A computer tablet will be used by the subjects at home during 14 days after their procedure to read emails, browse the internet and watch videos. Facial video recordings will be automatically acquired during these daily activities by the tablet device. The subject will be wearing an ECG patch during the follow-up period. The one-lead continuous ECG will be used as a reference to verify the presence of AF rhythm during facial video recordings. The primary aim of the study is to demonstrate the validity and robustness of the video-based technology to detect the presence of AF when facial videos are acquired by the patients at home.
Interventions
DIAGNOSTIC_TESTFacial video of detection of cardiac disease
This project proposes to evaluate a non-contact video recording technology to detect the presence of AF. The technology extracts a pulsatile signal by measuring the subtle variations in skin color of a patient's face (flushing) due to the variations of blood volume underneath the skin. The technology uses a standard web camera. This technique is videoplethysmography (VPG).
Sponsors
National Institutes of Health (NIH)
University of Rochester
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Men and women older than 18 years of age, * Medically-managed for symptomatic AF (persistent or paroxysmal), * In sinus rhythm after their ablation procedure, * In sinus rhythm after trans-thoracic electrical cardioversion.
Exclusion criteria
* Implanted with a device (pacemaker, CRT, ICD) and a ventricular pacing requirement superior or equal to 70%, * Known allergic reaction to adhesives or hydrogels or with a family history of adhesive skin allergies, * Unable to cooperate with the protocol due to dementia, psychological, or other related reason, * Refusing to sign the consent for participation, * Unable to operate the device such as blind patients. * Patients with Parkinson disease (or other central nervous system disorder/Tremor) who cannot record a stable video signal of their face. * The subject was previously enrolled in the current study. * The subject wears clothing covering the face, or uses facial makeup that will interfere with the quality of facial recordings. * Subject does not have internet access at home and lacks the technology to participate.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Area Under the Receiver Operating Characteristic Curve (AUC) for the Detection of Atrial Fibrillation | 2 weeks | The outcome measure is the AUC of the curve obtained from the ROC analysis. The curve is a combination of specificity and sensitivity of detecting atrial fibrillation. The range of the curve is 0.5-1.0 where 0.5 would be random detection of atrial fibrillation and 1.0 would be perfect performance. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| All Participants Facial video of detection of cardiac disease: This project proposes to evaluate a non-contact video recording technology to detect the presence of AF. The technology extracts a pulsatile signal by measuring the subtle variations in skin color of a patient's face (flushing) due to the variations of blood volume underneath the skin. The technology uses a standard web camera. This technique is videoplethysmography (VPG). | 255 |
| Total | 255 |
Withdrawals & dropouts
| Period | Reason | FG000 |
|---|---|---|
| Overall Study | Adverse Event | 4 |
| Overall Study | Withdrawal by Subject | 1 |
Baseline characteristics
| Characteristic | All Participants |
|---|---|
| Age, Continuous | 65.2 years STANDARD_DEVIATION 9.7 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 4 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 245 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 6 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants |
| Race (NIH/OMB) Black or African American | 2 Participants |
| Race (NIH/OMB) More than one race | 2 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 2 Participants |
| Race (NIH/OMB) White | 249 Participants |
| Region of Enrollment United States | 255 participants |
| Sex: Female, Male Female | 85 Participants |
| Sex: Female, Male Male | 170 Participants |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | 0 / 256 |
| other Total, other adverse events | 0 / 256 |
| serious Total, serious adverse events | 0 / 256 |
Outcome results
Primary
Area Under the Receiver Operating Characteristic Curve (AUC) for the Detection of Atrial Fibrillation
The outcome measure is the AUC of the curve obtained from the ROC analysis. The curve is a combination of specificity and sensitivity of detecting atrial fibrillation. The range of the curve is 0.5-1.0 where 0.5 would be random detection of atrial fibrillation and 1.0 would be perfect performance.
Time frame: 2 weeks
Population: We collected data for the AUC from 108 patients. We collected 977 recordings from these 108 patients.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| All Participants | Area Under the Receiver Operating Characteristic Curve (AUC) for the Detection of Atrial Fibrillation | 0.74 probability |