Presage HRV Model Card
THE FOLLOWING INFORMATION, AND PRESAGE'S SDK AND APP, ARE OFFERED FOR GENERAL WELLNESS AND INFORMATIONAL PURPOSES ONLY. NEITHER HAVE BEEN CLEARED BY THE FDA AND NEITHER MAY BE USED FOR MEDICAL DIAGNOSIS OR TREATMENT.
1. Model Details
Basic info: Presage vitals by video analysis generates heart rate variability metrics (RMSSD, SDNN, Baevsky Stress Index) from a video of a subject. Output frequency: one measurement per 60 seconds.
Organization: Presage Technologies
Model date: 2026-08-06
Model version: dc437d54
Model type: A proprietary computer vision and signal processing pipeline extracts a photoplethysmogram (PPG) from video of a subject, from which NN intervals are derived. HRV metrics reported: RMSSD = sqrt(mean((NNi+1 - NNi)2)), the root mean square of successive NN differences (ms); SDNN = sqrt(mean((NNi - mean(NN))2)), the standard deviation of NN intervals (ms); and Baevsky Stress Index = AMo / (2 × Mo × MxDMn), a measure of autonomic balance derived from the NN interval distribution.
License: The algorithm is currently proprietary, and licenses are granted with predefined agreement.
Contact: Questions can be sent to: support@presagetech.com
2. Intended Use
Model Uses
This HRV model was intended for use in analysis and non-diagnostic utility of HRV. It requires at least 60 consecutive seconds of unobstructed facial view acquired at 25+ fps. Only intended for participants with pulse rate 40-180 BPM.
Out-of-Scope Uses
As noted above. Not intended for diagnostic purposes, arrhythmia detection, or monitoring. Not intended for highly dynamic environments.
3. Validation
Reference Standard: Biopac MP160 3-lead ECG. R-peaks detected from ECG; NN intervals computed; HRV metrics derived from NN intervals over matched 60-second windows.
Comparison Methodology: HRV (RMSSD, SDNN, Baevsky) measurements from the camera-based system
were compared against time-aligned reference measurements. Ground truth signals were checked for
quality using labeled signal annotations; segments with poor signal quality were excluded from analysis.
4. Data Demographics
| Category |
Distribution |
| Total |
118 subjects, 418 videos |
| Camera (videos) |
e-con See3CAM CU27: 114, Logitech C920: 113, Samsung S24 Rear 24mm (tripod): 101, Samsung S24 Front 27mm (handheld): 90 |
| Sex (subjects) |
Female: 63, Male: 55 |
| Age Group (subjects) |
18-25: 34, 26-35: 39, 36-45: 17, 46-55: 15, 56-65: 10, 65+: 2 |
| Fitzpatrick (subjects) |
Type 1: 21, Type 2: 12, Type 3: 13, Type 4: 24, Type 5: 33, Type 6: 15 |
| BMI (subjects) |
Mean: 27.9, Range: 16.5–54.3, N=118 |
| Lighting (videos) |
Ring Light: 133, Lamp: 132, Window: 99 |
Reference standard: Biopac 3-lead ECG. Confidence values averaged across each 60-second period.
5. Data Provenance
Reference Instrumentation: Biopac research-grade physiological sensors:
ECG (for HRV ground truth).
Camera Devices Tested: Samsung S24 Rear 24mm (tripod), Samsung S24 Front 27mm (handheld),
e-con See3CAM CU27 (USB), Logitech C920 (tripod).
Average Camera Distance: Samsung S24/e-con/Logitech: 26", Samsung S24 handheld: 10"
Data Handling: All subject data is de-identified. Derived metrics and anonymized identifiers are
retained. Data is securely stored with access restricted to trained researchers.
6. Factors
The HRV model requires face detection for photoplethysmographic signal extraction.
- User: Face must be within angular detection parameters, visible and illuminated.
- Environment: Developed on videos with various lighting and motion conditions.
These factors can affect model performance:
- Lighting: Affects plethysmogram quality and thus HRV accuracy.
- User: Skin tone and facial hair may impact signal quality.
- Motion: High user motion degrades the plethysmographic signal.
- Chewing gum can degrade performance.
Lighting Conditions Tested:
- Ring Light: Direct, front-facing illumination providing consistent lighting on the subject's face.
- Window: Natural sunlight entering through a window that illuminates the subject's face from the side rather than directly head-on. Lighting intensity varies depending on weather conditions and the time of day when the measurement is collected.
- Lamp: A room lighting lamp aimed at the ceiling and not directed at the participant, to mimic living room-like environments.
Other factors:
- Instrumentation: Camera parameter locking (exposure, gain, white balance) is critical for HRV accuracy.
7. Metrics
(at 80% Return Rate, Confidence >= 79)
- Mean NN MAE: 3.12 ms
- Mean NN RMSE: 4.41 ms
- RMSSD MAE: 10.19 ms
- RMSSD RMSE: 13.74 ms
- SDNN MAE: 5.35 ms
- SDNN RMSE: 7.72 ms
8. Quantitative Analysis
Computed vs Ground Truth at Confidence Thresholds
Bland-Altman Plot
Confidence Lookup Table
| Confidence >= |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
Return Rate (%) |
N (samples) |
| 0 |
4.49 |
8.71 |
14.33 |
21.32 |
8.23 |
14.13 |
100.0 |
8067 |
| 10 |
4.49 |
8.71 |
14.33 |
21.32 |
8.23 |
14.13 |
100.0 |
8067 |
| 20 |
3.54 |
5.31 |
12.29 |
17.07 |
6.87 |
10.95 |
95.7 |
7720 |
| 30 |
3.52 |
5.29 |
12.05 |
16.74 |
6.71 |
10.77 |
94.6 |
7633 |
| 40 |
3.44 |
5.15 |
11.79 |
16.22 |
6.52 |
10.27 |
93.6 |
7551 |
| 50 |
3.36 |
5.00 |
11.33 |
15.46 |
6.17 |
9.41 |
91.4 |
7376 |
| 60 |
3.25 |
4.80 |
10.61 |
14.22 |
5.67 |
8.25 |
87.6 |
7063 |
| 65 |
3.22 |
4.76 |
10.44 |
14.04 |
5.57 |
8.14 |
85.4 |
6889 |
| 70 |
3.21 |
4.73 |
10.39 |
14.00 |
5.54 |
8.11 |
83.5 |
6733 |
| 75 |
3.14 |
4.48 |
10.37 |
13.99 |
5.46 |
7.92 |
81.7 |
6592 |
| 79 |
3.12 |
4.41 |
10.19 |
13.74 |
5.35 |
7.72 |
80.0 |
6432 |
| 80 |
3.10 |
4.38 |
10.16 |
13.68 |
5.33 |
7.68 |
79.3 |
6401 |
| 85 |
3.00 |
4.22 |
9.75 |
12.96 |
5.06 |
7.07 |
76.1 |
6140 |
| 90 |
2.91 |
4.08 |
9.59 |
12.66 |
4.91 |
6.80 |
70.1 |
5654 |
| 95 |
2.89 |
4.05 |
8.96 |
11.89 |
4.54 |
6.25 |
62.1 |
5007 |
| 96 |
2.87 |
4.04 |
8.72 |
11.56 |
4.45 |
6.16 |
59.7 |
4813 |
| 97 |
2.81 |
3.93 |
8.62 |
11.54 |
4.33 |
5.88 |
55.9 |
4506 |
| 98 |
2.83 |
3.98 |
8.33 |
11.32 |
4.14 |
5.70 |
50.7 |
4089 |
| 99 |
2.91 |
4.09 |
7.79 |
10.95 |
3.89 |
5.46 |
41.2 |
3326 |
Performance
(at 80% Return Rate, Confidence >= 79)
By Camera Type
| Camera Type |
N (samples) |
Return Rate (%) |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
| Samsung S24 Rear 24mm (tripod) |
1974 |
89.0 |
3.41 |
4.92 |
10.57 |
14.47 |
5.98 |
8.85 |
| Samsung S24 Front 27mm (handheld) |
426 |
49.2 |
2.96 |
4.08 |
10.34 |
12.42 |
5.16 |
8.07 |
| e-con See3CAM CU27 |
1886 |
76.3 |
3.10 |
4.27 |
9.67 |
13.45 |
5.27 |
7.56 |
| Logitech C920 |
2146 |
85.4 |
2.89 |
4.10 |
10.28 |
13.55 |
4.88 |
6.59 |
By Fitzpatrick Skin Type
| Fitzpatrick |
N (samples) |
Return Rate (%) |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
| Type I |
1269 |
95.6 |
3.67 |
5.27 |
8.34 |
10.78 |
3.37 |
4.34 |
| Type II |
878 |
94.9 |
2.44 |
3.14 |
6.87 |
9.01 |
4.13 |
5.50 |
| Type III |
940 |
92.3 |
4.11 |
5.32 |
8.09 |
10.66 |
4.94 |
7.77 |
| Type IV |
1439 |
86.3 |
2.95 |
4.30 |
9.68 |
12.68 |
5.27 |
6.69 |
| Type V |
1557 |
68.3 |
2.63 |
3.65 |
13.01 |
16.46 |
6.82 |
9.33 |
| Type VI |
349 |
41.0 |
3.00 |
4.59 |
20.52 |
25.99 |
10.53 |
14.77 |
| Fitzpatrick Type VI: Breakdown by Lighting Condition |
| Ring Light |
165 |
57.7 |
2.18 |
2.90 |
18.95 |
19.94 |
8.24 |
9.57 |
| Window |
17 |
6.3 |
6.24 |
7.11 |
25.10 |
27.55 |
6.70 |
7.08 |
| Lamp |
129 |
50.2 |
3.77 |
6.06 |
24.76 |
34.31 |
14.95 |
21.18 |
Note: Types I-V are composites across all lighting conditions (Ring Light, Window, Lamp).
By Sex
| Sex |
N (samples) |
Return Rate (%) |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
| Male |
3309 |
83.7 |
3.32 |
4.75 |
8.53 |
11.23 |
4.68 |
6.41 |
| Female |
3123 |
75.9 |
2.91 |
4.03 |
11.96 |
15.97 |
6.06 |
8.90 |
By Age Group
| Age Group |
N (samples) |
Return Rate (%) |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
| 18-25 |
2036 |
85.3 |
4.12 |
5.50 |
8.97 |
12.53 |
5.31 |
7.96 |
| 26-35 |
1954 |
76.0 |
2.96 |
3.84 |
7.57 |
11.34 |
4.10 |
6.91 |
| 36-45 |
862 |
75.0 |
2.69 |
3.68 |
13.34 |
15.73 |
5.86 |
7.51 |
| 46-55 |
935 |
85.5 |
1.78 |
2.49 |
14.16 |
15.91 |
7.16 |
8.55 |
| 56-65 |
571 |
84.1 |
2.64 |
4.80 |
9.86 |
13.01 |
4.78 |
5.75 |
| 65+ |
14 |
11.8 |
6.13 |
7.19 |
30.32 |
30.35 |
7.25 |
7.57 |
By Lighting Type
| Lighting |
N (samples) |
Return Rate (%) |
Mean NN MAE (ms) |
Mean NN RMSE (ms) |
RMSSD MAE (ms) |
RMSSD RMSE (ms) |
SDNN MAE (ms) |
SDNN RMSE (ms) |
| Lamp |
2072 |
80.5 |
3.39 |
4.92 |
9.58 |
13.74 |
5.26 |
8.50 |
| Ring Light |
2261 |
83.7 |
2.89 |
3.89 |
11.17 |
14.73 |
5.51 |
7.71 |
| Window |
1228 |
66.6 |
3.50 |
5.01 |
8.92 |
11.35 |
4.80 |
6.02 |
9. Fairness & Equity
Bias Assessment Methodology: Performance is stratified by Fitzpatrick skin type (I-VI),
sex, camera type, and age group. Per-group metrics and Confidence averages are reported in the
Quantitative Analysis tables above.
10. Ethical Considerations
As a remote sensing device, the risks posed to the subjects in the trial are minimal, including the association of each subject with corresponding biometric data. Mitigation of these risks include de-identifying all subject data, including videos, prior to saving it. Additionally, all data is securely stored with access to a select number of trained researchers.
The model is not intended for human life-critical decisions, diagnostics or prognostication.
11. Limitations and Tradeoffs
- Requires at least 60 seconds of uninterrupted facial view
- Works best in stationary conditions
- HRV metrics are sensitive to motion artifacts
- Currently only measures a single subject at a time
