Oura Ring Gen 4 sensor data, not clinical measurementsN=1 case study, not validated for clinical decisionsHEV diagnosed Mar 18; Day 109 post-ruxolitinibMore
Consumer wearable data can support exploratory review only. The HEV diagnosis, temporally confounded with treatment start, remains a material confounder.

Anomaly Detection

ML-powered analysis: Matrix Profile, Isolation Forest, LSTM Autoencoder, SPC, tsfresh+DBSCAN
Generated 2026-07-02 19:36 · Post-HSCT Patient · Data Jan 08, 2026 to Jul 02, 2026 · Day 109 post-treatment
Anomaly Days
Elevated
18.0
Above 90th percentile
Feb 9 Detection
Detected
2/5
40% method agreement (N=1)
High Severity
Detected
116.0
Score >= 0.7
Moderate
Detected
34.0
Score 0.4-0.7
52% higher mean anomaly score after ruxolitinib (109 days)
Data period: 2026-01-08 to 2026-07-02 (178 days) · Known acute event: 2026-02-09
REPORT SECTION

Validation: February 9 Event

Methods Detected
Low
2/5
Method Agreement (N=1)
Weak
40%

Detection results per method:

Matrix ProfileIsolation ForestLstm AutoencoderSpcTsfresh

The Feb 8-9, 2026 event is a clinically confirmed acute episode (HR spike to 113 bpm mean, RMSSD drop to 5.0 ms, readiness score 29). All anomaly detectors are validated against this event.

REPORT SECTION

Anomaly Timeline

REPORT SECTION

Statistical Process Control (SPC)

REPORT SECTION

LSTM Autoencoder - Training Curve

REPORT SECTION

Top Anomalies (Ensemble Scoring)

#DateEnsemble ScoreRMSSD (ms) HR (bpm)EfficiencyReadinessMethods (score > 0.5)
12026-02-09 **1.2225.011076%292/5
22026-04-081.1989.111267%502/5
32026-05-311.19835.983-%632/5
42026-04-091.11220.284-%-2/5
52026-05-301.11023.59173%642/5
62026-06-041.10326.08458%652/5
72026-04-051.0899.411169%662/5
82026-05-211.07325.49888%602/5
92026-05-181.07220.18682%692/5
102026-01-311.0576.510779%482/5
112026-04-061.0526.911481%492/5
122026-06-061.05024.86973%792/5
132026-05-281.03455.49583%702/5
142026-06-201.02918.08963%622/5
152026-04-241.02646.58583%802/5

** = known acute event. Methods (score > 0.5) is a soft support count and is not the same as strict Feb 9 detection validation above.

REPORT SECTION

Clinical Interpretation

Summary

Five independent anomaly detection algorithms were run retrospectively on 178 days of biometric data from the Oura Ring (Jan-Mar 2026).

Validation Against Known Event

2 of 5 methods detected the acute event on Feb 8-9, 2026. The event ranks as #1 in ensemble anomaly scoring.

This event showed dramatic HRV collapse (RMSSD 5.0 ms), mean heart rate 113.9 bpm during sleep, and readiness score 29/100 with recovery index 4/100 - consistent with acute autonomic destabilization.

Top 5 Anomalies

  1. 2026-02-09
  2. 2026-04-08
  3. 2026-05-31
  4. 2026-04-09
  5. 2026-05-30

Clinical Relevance

Baseline RMSSD ~10 ms is critically low (parasympathetic deficiency threshold (Kleiger 1987 / Bigger 1992): 15 ms, population median: 49 ms, typical post-HSCT: 25-40 ms). Any further decline represents severe autonomic stress. Mean nightly heart rate >89 bpm meets IST criteria (>90 bpm at rest).

Limitations

  • Only 1 known positive event for validation (no specificity/PPV calculation possible)
  • LSTM autoencoder has limited training data (178 days) - results should be interpreted cautiously
  • Oura Ring measurements are not medical devices and have inherent noise
  • tsfresh analysis depends on the number of HR data points per night (varies significantly)
REPORT SECTION

Method Summary

MethodRuntimeFeb 9 DetectedDetails
Matrix Profile0sNostumpy not installed
Isolation Forest0.33sYes18 anomalies / 178 days
Lstm Autoencoder0sNotorch not installed
Spc0.01sYesShewhart + CUSUM + EWMA
Tsfresh0sNotsfresh not installed
Data window: January 08 to July 02, 2026 · Post-treatment: day 109 on ruxolitinib · Generated: 2026-07-02 19:36
Oura Ring Gen 4 consumer wearable data; N=1 exploratory case study; not a medical device. MIT License · © 2026 The Educational Equality Institute · Oura® is a registered trademark of Oura Health Oy.