Physiological biomarkers

The autonomic nervous system, read from wearables.

Heart rate variability from RR intervals or PPG, electrodermal activity, sleep and the rest-activity rhythm, with the Task Force standards and published nonparametric methods.

For research and clinician review. Not a diagnosis.

medera · /v1/physiological/analyze
sample
Heartbeatsmean heart rate 69.8 bpm869 ms910 ms887 ms825 ms831 ms876 ms873 msRR series · first 64 of 350 beats800860920Lomb–Scargle spectrumpower, square-root scale00.10.20.30.4 HzLFHFPoincaré plotRRₙ (ms)RRₙ₊₁SD2 35.66 msSD1 24.18 msRMSSD34.15msSDNN30.47msHF power641ms²LF/HF ratio0.396
Explained

Physiological, step by step.

1:06 with on-screen captions. Every number in it is the engine's output for the published sample.

Read the transcript
  1. 01 · The body keeps time

    Heart rhythm, skin conductance, sleep and activity carry signals of arousal and regulation.

  2. 02 · Beats to intervals

    Each heartbeat is located, and the time between beats becomes an RR interval.

    Ectopic and out-of-range beats are screened out. At least 80% must be normal.

  3. 03 · Variability

    Five minutes of intervals form a series. Its spread is SDNN; beat-to-beat change is RMSSD.

    The Poincaré plot shows both at once: SD1 across the line, SD2 along it.

  4. 04 · The spectrum

    A Lomb–Scargle periodogram reads the uneven series directly, with no resampling.

    High-frequency power follows breathing and vagal tone; low-frequency power is slower.

  5. 05 · What comes back

    For this series the API returns RMSSD 34.15 ms, SDNN 30.47 ms and HF power 641 ms².

  6. 06 · Sleep and rhythm

    Sleep and the daily rest-activity rhythm are measured from the same kind of numbers.

    A week here: 7.44 hours of sleep, 89.3% efficiency, interdaily stability 0.982.

  7. 07 · Quality gates

    A band exists only when the recording is long enough: HF 1 minute, LF 2, VLF 5.

    Too short, and the value is null, with the reason in the response.

  8. 08 · One request

    Send numbers, not vendor files: any wearable that exports intervals, PPG, sleep or activity works.

    Every signal is optional. Send what you have.

  9. 09 · Recap

    Task Force standards, published methods, and a value only when the data supports it.

What goes in

Inputs, and only those.

Series are processed in memory and discarded when the response is sent.

  • 01RR or inter-beat intervals in milliseconds, or a raw PPG signal sampled at 20 Hz or more
  • 02Skin conductance in microsiemens at 1 Hz or more
  • 03Nightly sleep summaries, or sleep and wake epochs for the Sleep Regularity Index
  • 04Hourly activity counts or daily steps
How it is measured

Published methods, in order.

Each step names the method behind it. The catalog lists the references for every measure.

  1. 01Clean the beats

    Ectopic and out-of-range intervals are screened. A recording needs 80% normal intervals before anything is reported.

  2. 02Time domain

    Mean heart rate, SDNN, RMSSD and pNN50 by the 1996 Task Force definitions, plus Poincaré SD1 and SD2.

  3. 03Frequency domain

    A Lomb–Scargle spectrum of the uneven RR series gives VLF, LF and HF power, each only when the recording is long enough.

  4. 04Electrodermal activity

    Tonic skin conductance level and non-specific responses per minute, with the 0.01 to 0.05 µS amplitude criterion.

  5. 05Sleep and rhythm

    Sleep continuity, night-to-night variability, the Sleep Regularity Index, and interdaily stability, intradaily variability and relative amplitude.

What comes out

The biomarkers, defined.

Every measure links to its catalog entry: definition, method, quality gate, published associations and references.

BiomarkerUnitWhat it measuresPublished associations
Cardiac
Mean heart ratemeanHrBpmbpm60,000 divided by the mean normal-to-normal interval.RDoC: Arousal and regulatory systemsDepression↑MAPost-traumatic stress↑MA
Mean NN intervalmeanNnMsmsMean normal-to-normal interval.Descriptive
SDNNsdnnMsmsStandard deviation of all NN intervals.RDoC: Arousal and regulatory systemsDepression↓MAPost-traumatic stress↓MA
RMSSDrmssdMsmsSquare root of the mean squared differences between adjacent NN intervals.RDoC: Arousal and regulatory systemsDepression↓MAAnxiety↓MAPost-traumatic stress↓MA+2
SDSDsdsdMsmsStandard deviation of differences between adjacent NN intervals.Descriptive
pNN50pnn50Pct%NN50 count divided by the total number of NN intervals.Descriptive
Poincaré SD1sd1MsmsShort-term Poincaré dispersion, SDSD/√2.Descriptive
Poincaré SD2sd2MsmsLong-term Poincaré dispersion, √(2·SDNN² − ½·SDSD²).Descriptive
VLF powervlfPowerMs2ms²Power 0.0033–0.04 Hz.Descriptive
LF powerlfPowerMs2ms²Power 0.04–0.15 Hz.Depression~MAPost-traumatic stress~MA
HF powerhfPowerMs2ms²Power 0.15–0.40 Hz (respiratory sinus arrhythmia).RDoC: Arousal and regulatory systemsDepression↓MAAnxiety↓MAPost-traumatic stress↓MA+2
LF/HF ratiolfHfRatioratioLF power over HF power.RDoC: Arousal and regulatory systemsDepression↑MAPost-traumatic stress↑MA
LF (normalized units)lfNun.u.100·LF/(LF+HF).Descriptive
HF (normalized units)hfNun.u.100·HF/(LF+HF).Descriptive
Electrodermal
Skin conductance levelsclMeanUsµSMean tonic skin conductance level.Depression↓RPost-traumatic stress↑MASuicide risk~
Non-specific SCRsscrPerMin/minSpontaneous skin conductance responses per minute.Depression~R
SCR amplitudescrAmplitudeMeanUsµSMean amplitude of the counted responses.Depression↓RBipolar↓SSuicide risk~
Sleep
Total sleep timetotalSleepTimeHhMean nightly total sleep time.Depression↓MAAnxiety↓MAPost-traumatic stress↓MA+2
Sleep efficiencysleepEfficiencyPct%Mean share of time in bed spent asleep.RDoC: Arousal and regulatory systemsDepression↓MAAnxiety↓MAPost-traumatic stress↓MA+1
Sleep onset latencysleepOnsetLatencyMinminMean time to fall asleep.RDoC: Arousal and regulatory systemsDepression↑MABipolar↑MASchizophrenia↑MA
Sleep timing variabilitysleepMidpointSdMinminNight-to-night standard deviation of the sleep midpoint (circular).RDoC: Arousal and regulatory systemsSuicide risk↑S
Sleep duration variabilitytotalSleepTimeSdMinminNight-to-night standard deviation of total sleep time.RDoC: Arousal and regulatory systemsBipolar↑MASchizophrenia↑MA
Sleep Regularity IndexsleepRegularityIndexindex (−100 to 100)Probability of being in the same sleep/wake state 24 h apart, rescaled to −100…100.RDoC: Arousal and regulatory systemsDepression↓SAnxiety↓S
Rest-activity
Overall activityactivityMeanCountsPerHourcounts/hMean hourly activity.RDoC: Sensorimotor systemsDepression↓MABipolar↓MASchizophrenia↓MA
Daily stepsdailyStepsMeansteps/dayMean daily step count.RDoC: Sensorimotor systemsDescriptive
M10m10counts/hMean activity of the most active 10 consecutive hours of the average day.Descriptive
L5l5counts/hMean activity of the least active 5 consecutive hours of the average day.Descriptive
Relative amplituderelativeAmplituderatio(M10 − L5)/(M10 + L5): strength of the rest-activity rhythm.RDoC: Arousal and regulatory systemsDepression↓RBipolar~
Interdaily stabilityinterdailyStability0–1How closely each day follows the average 24-h profile.RDoC: Arousal and regulatory systemsDepression↓MA
Intradaily variabilityintradailyVariability0–2Fragmentation of the rhythm from hour to hour.Bipolar=MASchizophrenia=MA

Associations are group-level findings from the literature (↑ higher, ↓ lower; MA meta-analysis, R replicated, S single study), never a diagnosis. Compare a person with themselves over time.

Quality gates

Null, never a guess.

When the data cannot support a measure, the value is null and the response says why in quality.flags.

  • HRV needs at least 60 s and 80% normal intervals; LF needs 2 min and VLF 5 min.
  • EDA needs at least 60 s of signal.
  • Sleep means need 3 nights; variability needs 5.
  • Rest-activity measures need 3 whole days of hourly data.
POST /v1/physiological/analyze

One request. Try it now.

Each example downloads the published sample and sends it with the public sandbox key. The sandbox takes synthetic data only.

# The published sample request (synthetic, no real person's data)
curl -s https://www.medera.info/api/v1/samples/physiological -o physiological.json

curl -s https://www.medera.info/api/v1/physiological/analyze \
  -H "Authorization: Bearer mk_sandbox_medera_public" \
  -H "Content-Type: application/json" \
  --data @physiological.json
POST/v1/physiological/analyze
Sample request

Sends the synthetic sample to the live sandbox with the public key and shows the full response. Nothing personal is sent and nothing is stored.

Response for the sample · 200
{
  "object": "physiological.analysis",
  "api_version": "v1",
  "mode": "sandbox",
  "engine": {
    "name": "medera-inference",
    "version": "1.0.0",
    "vocal": "BH-ASR voice layer 1.2.0 (TypeScript port, number-for-number with the Python reference)"
  },
  "intended_use": "Measured biomarkers and within-person change for research and clinician review. Not a diagnosis and not a medical device.",
  "biomarkers": {
    "meanHrBpm": 69.8,
    "meanNnMs": 859.6,
    "sdnnMs": 30.47,
    "rmssdMs": 34.15,
    "sdsdMs": 34.2,
    "pnn50Pct": 13.14,
    "sd1Ms": 24.18,
    "sd2Ms": 35.66,
    "vlfPowerMs2": 4.8,
    "lfPowerMs2": 253.7,
    "hfPowerMs2": 641,
    "lfHfRatio": 0.396,
    "lfNu": 28.4,
    "hfNu": 71.6,
    "sclMeanUs": null,
    "scrPerMin": null,
    "scrAmplitudeMeanUs": null,
    "totalSleepTimeH": 7.44,
    "sleepEfficiencyPct": 89.3,
    "sleepOnsetLatencyMin": 20.1,
    "sleepMidpointSdMin": 20.5,
    "totalSleepTimeSdMin": 16.6,
    "sleepRegularityIndex": null,
    "activityMeanCountsPerHour": 736.36,
    "dailyStepsMean": 6799,
    "m10": 1350.64,
    "l5": 39.6,
    "relativeAmplitude": 0.943,
    "interdailyStability": 0.982,
    "intradailyVariability": 0.267
  },
  "quality": {
    "heart": {
      "beats": 350,
      "normal_fraction": 1,
      "duration_s": 300.9,
      "flags": [],
      "source": "rr"
    },
    "eda": null,
    "sleep": {
      "nights": 7,
      "flags": []
    },
    "activity": {
      "days": 7,
      "flags": []
    }
  }
}
Computed by the engine for the published sample. Each call also carries an id and a created timestamp. Full reference
FAQ

Physiological, answered.

Still wondering about something?

A real person will answer, usually the same day.

Any device or SDK that exports RR intervals, PPG, skin conductance, sleep or activity data. The API takes numbers, not vendor files.

RR intervals are unevenly spaced in time. Resampling before an FFT distorts the spectrum; the Lomb–Scargle periodogram works on the uneven series directly.

Combine them

Better together, compared over time.

Send this modality with the others in one call to /v1/analyze, or pass the biomarkers to the inference engine.

Start measuring.

Sandbox today. Production under agreement.

Or make a first request with the sandbox key.