Exploratory N=1 Physiological State-Space Model
Bayesian Cardiovascular Digital Twin
This page compresses 79 wearable days into five latent physiological states, combining a Kalman smoother with UKF nonlinear dynamics to track recovery, instability, and treatment response. Use it for relative trajectory tracking and model diagnostics, not for diagnosis or causal proof.
Generated 2026-03-27 07:10 · 2026-01-04 to 2026-03-23 · Post-drug window: 8 days
5 latent states5 wearable inputsAcute event 2026-02-09Ruxolitinib 2026-03-16HEV Dx 2026-03-18Post-HSCT Patient
Strongest modeled shift
+1.07 SD
Cardiac Reserve · Improved · p=0.040 · exploratory
Best short-term fit
R-sq 0.353
HRV (RMSSD) · RMSE 2.08 ms
Residual checks
5/5
Ljung-Box p > 0.05 across modeled sensors
Post-drug window
8 days
Short window; HEV diagnosed 2026-03-18
79
2026-01-04 to 2026-03-23
Post-Drug Days
Insufficient8
Since 2026-03-16
Innovation Alerts
Elevated1
22 total across the modeled window
Avg Sensor Coverage
Normal86.6%
Lowest: SpO2 (75%)
5/5
Ljung-Box p>0.05 across modeled sensors
Read This First
If someone opens only one part of this page, it should be this block. It summarizes the signal, the diagnostics that support the model, and the reasons the interpretation remains exploratory.
What the model suggests
- Strongest modeled post-drug shift: Cardiac Reserve +1.07 SD (p=0.040, exploratory).
- Best one-step predictive stream: HRV (RMSSD) with R-sq 0.353 and RMSE 2.08 ms.
- Recent instability is limited: 1 innovation alerts in the last 7 days (22 total across the modeled window).
What supports the model
- Residual diagnostics pass for 5/5 modeled sensors.
- Average sensor coverage is 86.6%; lowest coverage is SpO2 at 75%.
- HRV (RMSSD) contributes the largest information share at 32.7%.
Why caution is still needed
- This is a single-patient (N=1) exploratory model, not a validated clinical instrument.
- The post-drug window is only 8 days, which is too short for strong treatment claims.
- HEV diagnosed 2026-03-18 may confound late-March shifts after ruxolitinib started on 2026-03-16.
Latent State Trajectories
The Kalman smoother estimates 5 latent physiological states from noisy, intermittent sensor data. Solid lines show the smoothed posterior mean; shaded bands show 95% credible intervals. Dotted lines overlay the UKF estimates for comparison. Vertical markers indicate the Feb 9 acute event, Mar 16 ruxolitinib start, and Mar 18 HEV diagnosis.
Ruxolitinib Drug Response (started 2026-03-16)
Latent-state shifts are standardized, so the pre/post comparison shows magnitude rather than raw clinical units. Use this block to gauge which modeled subsystems moved most after treatment began. The post-drug window is short, and HEV diagnosed on 2026-03-18 may confound late-March movement.
Largest shiftCardiac Reserve+1.07 SD · Improved · p=0.040
Fastest fitted responseSleep Quality0.3 days to reach modeled equilibrium
Positive shifts indicate a higher modeled state load after treatment start; time constants estimate how quickly the post-drug response stabilized. P-values here are unadjusted and should be treated as descriptive, not confirmatory.
| State | Pre-drug Mean | Post-drug Mean | Shift (SD) | How to read direction | Direction | p-value | Time Constant |
|---|
| Autonomic Tone | -0.08 | 0.27 | +0.474 | Higher = stronger vagal/recovery signal | Improved (pre/post drug) | p=0.244 | 3.2 days |
| Cardiac Reserve | -0.44 | 0.54 | +1.071 | Higher = better modeled cardiovascular resilience | Improved (pre/post drug) | p=0.040 | 5.4 days |
| Circadian Phase | 0.96 | 0.07 | -1.659 | Direction is descriptive; interpret with timing context | Stable (pre/post drug) | p<0.001 | 4.8 days |
| Inflammation Level | -0.56 | 0.09 | +0.673 | Higher = worse modeled inflammatory load | Worsened (pre/post drug) | p=0.008 | 4.9 days |
| Sleep Quality | -0.14 | 0.37 | +0.725 | Higher = better modeled sleep/recovery signal | Improved (pre/post drug) | p<0.001 | 0.3 days |
Observations vs Model Estimates
Raw sensor observations (dots) overlaid with the model's filtered estimates (lines). Good model fit is indicated by the line tracking the dots closely. Vertical markers indicate the Feb 9 acute event, Mar 16 ruxolitinib start, and Mar 18 HEV diagnosis.
Prediction Performance
One-step-ahead residuals show how quickly the model tracks changes in physiology. Higher R-sq and lower RMSE indicate the latent-state model is anticipating the next observation well. Vertical markers are shown on the time-series subplots for the acute event, treatment start, and HEV diagnosis.
Per-sensor forecast quality
HRV (RMSSD)
R-sq 0.353
RMSE 2.08 ms
Heart Rate
R-sq 0.159
RMSE 5.97 bpm
SpO2
R-sq 0.027
RMSE 0.54 %
Sleep Efficiency
R-sq 0.020
RMSE 3.33 %
Temperature Deviation
R-sq -0.191
RMSE 0.27 C
Innovation alerts (22 total, 1 in last 7 days)
2026-03-23: HRV (RMSSD) deviation = 3.3 sigma
KF vs UKF Comparison
The Unscented Kalman Filter uses nonlinear circadian dynamics and exponential autonomic decay with sigma-point propagation (no Jacobian needed). Large differences from the linear KF suggest significant nonlinear dynamics.
Multi-Modal Sensor Fusion Quality
This block shows which wearable streams are carrying the latent-state estimates and whether the residuals still contain structure the model failed to absorb.
Sensor contribution to state estimation
HRV (RMSSD)
32.7%
share of the fused state-estimation signal
Heart Rate
22.5%
share of the fused state-estimation signal
Sleep Efficiency
16.6%
share of the fused state-estimation signal
Temperature Deviation
14.1%
share of the fused state-estimation signal
SpO2
14.0%
share of the fused state-estimation signal
Residual diagnostics
White noise residuals (Ljung-Box p > 0.05) indicate the model captures temporal structure well. Significant autocorrelation suggests model misspecification for that sensor.
Disclaimer: This is a computational model for research and self-monitoring purposes only. It is NOT a medical device and should NOT be used for clinical decision-making. The Oura Ring is a consumer wearable; its measurements have known limitations in accuracy. All state estimates are model-dependent and should be interpreted with appropriate uncertainty. Consult qualified healthcare professionals for medical decisions.