Get API key

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: 2026-11-06

Model version: d205a6f0

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, 413 videos
Camera (videos) Logitech C920: 113, e-con See3CAM CU27: 111, Samsung S24 Rear 24mm (tripod): 99, Samsung S24 Front 27mm (handheld): 90
Sex (subjects) Female: 63, Male: 55
Age Group (subjects) 18-25: 33, 26-35: 39, 36-45: 18, 46-55: 13, 56-65: 12, 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: 131, Lamp: 130, Window: 98

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)

  1. Mean NN MAE: 3.16 ms
  2. Mean NN RMSE: 4.55 ms
  3. RMSSD MAE: 10.17 ms
  4. RMSSD RMSE: 13.72 ms
  5. SDNN MAE: 5.25 ms
  6. SDNN RMSE: 7.58 ms

8. Quantitative Analysis

Computed vs Ground Truth at Confidence Thresholds

HRV Mean NN in 60-second windows, computed vs. ground truth, for the top 50%, the top 80% and all readings by confidence, colored by cameraHRV SDNN in 60-second windows, computed vs. ground truth, for the top 50%, the top 80% and all readings by confidence, colored by cameraHRV RMSSD in 60-second windows, computed vs. ground truth, for the top 50%, the top 80% and all readings by confidence, colored by camera

Bland-Altman Plot

HRV Bland-Altman plots at an 80% return rate for Mean NN, RMSSD and SDNN

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 4.60 9.04 14.27 21.37 8.31 14.79 100.0 8034
10 4.60 9.04 14.27 21.37 8.31 14.79 100.0 8034
20 3.67 5.80 12.44 17.49 6.93 11.40 96.4 7741
30 3.57 5.50 12.04 16.64 6.64 10.71 95.2 7648
40 3.45 5.17 11.68 15.98 6.40 10.08 93.8 7537
50 3.36 4.97 11.19 15.20 6.06 9.24 91.4 7342
60 3.30 4.86 10.57 14.22 5.63 8.18 87.4 7024
65 3.24 4.75 10.39 14.03 5.47 7.88 85.8 6894
70 3.23 4.74 10.33 13.98 5.43 7.85 83.8 6736
75 3.23 4.75 10.31 13.96 5.41 7.86 82.2 6603
79 3.16 4.55 10.17 13.72 5.25 7.58 80.0 6438
80 3.12 4.42 10.14 13.65 5.18 7.43 79.5 6386
85 3.06 4.32 9.91 13.28 5.05 7.09 76.6 6155
90 2.89 4.03 9.65 12.86 4.87 6.72 70.1 5633
95 2.91 4.06 9.03 12.00 4.61 6.34 63.0 5059
96 2.88 4.05 8.72 11.56 4.49 6.15 60.3 4847
97 2.81 3.93 8.53 11.40 4.27 5.73 55.6 4465
98 2.84 3.99 8.26 11.20 4.13 5.62 50.1 4024
99 2.94 4.12 7.63 10.67 3.84 5.32 41.1 3298

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) 1960 88.0 3.38 4.90 10.20 14.19 5.74 8.58
Samsung S24 Front 27mm (handheld) 454 53.0 3.31 5.50 10.98 13.58 5.26 8.48
e-con See3CAM CU27 1882 77.2 3.17 4.39 9.70 13.41 5.11 7.20
Logitech C920 2142 85.3 2.91 4.11 10.39 13.60 4.92 6.68

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 1275 96.1 3.87 5.86 8.24 10.75 3.50 4.77
Type II 873 95.0 2.45 3.16 6.93 8.92 4.09 5.37
Type III 959 94.4 4.08 5.30 8.41 11.20 4.93 7.73
Type IV 1441 86.0 2.91 4.27 9.62 12.56 5.16 6.54
Type V 1540 68.1 2.62 3.56 12.71 16.17 6.44 8.71
Type VI 350 42.0 3.09 4.78 21.20 26.47 10.57 15.14
Fitzpatrick Type VI: Breakdown by Lighting Condition
Ring Light 165 56.1 2.20 2.90 19.06 20.17 8.23 9.65
Window 17 6.3 6.25 7.12 25.11 27.55 6.70 7.08
Lamp 130 55.6 3.99 6.44 26.42 35.05 15.03 21.78

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 3317 84.6 3.34 4.77 8.61 11.39 4.68 6.44
Female 3121 75.9 2.95 4.30 11.84 15.83 5.86 8.63

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 1953 84.8 4.33 5.97 8.53 12.03 5.15 7.78
26-35 2044 79.0 2.89 3.75 7.76 11.55 4.11 6.81
36-45 865 73.7 2.79 3.83 12.99 15.34 5.71 7.30
46-55 772 82.1 1.69 2.09 15.31 16.79 6.86 8.13
56-65 728 87.7 2.45 4.37 10.42 13.41 5.38 6.42
65+ 14 10.7 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 2085 81.2 3.55 5.38 9.39 13.53 5.19 8.57
Ring Light 2250 83.7 2.82 3.75 10.99 14.62 5.17 7.14
Window 1234 67.4 3.47 4.97 9.08 11.43 4.91 6.13

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

Download the full model card HTML, 9.6 MB