Datasets:
scenario_id string | pressure float64 | buffer_capacity float64 | coupling_strength float64 | oscillation_amplitude float64 | oscillation_frequency float64 | trajectory_drift float64 | label_oscillatory_instability int64 |
|---|---|---|---|---|---|---|---|
coi_train_001 | 0.44 | 0.76 | 0.36 | 0.08 | 0.1 | -0.02 | 0 |
coi_train_002 | 0.48 | 0.72 | 0.39 | 0.1 | 0.12 | 0.01 | 0 |
coi_train_003 | 0.52 | 0.69 | 0.42 | 0.14 | 0.16 | 0.03 | 0 |
coi_train_004 | 0.57 | 0.64 | 0.46 | 0.22 | 0.24 | 0.06 | 1 |
coi_train_005 | 0.61 | 0.6 | 0.49 | 0.28 | 0.3 | 0.09 | 1 |
coi_train_006 | 0.66 | 0.55 | 0.53 | 0.33 | 0.36 | 0.12 | 1 |
coi_train_007 | 0.7 | 0.5 | 0.57 | 0.38 | 0.4 | 0.15 | 1 |
coi_train_008 | 0.75 | 0.45 | 0.62 | 0.42 | 0.45 | 0.18 | 1 |
coi_train_009 | 0.47 | 0.74 | 0.38 | 0.09 | 0.11 | -0.01 | 0 |
coi_train_010 | 0.53 | 0.68 | 0.43 | 0.16 | 0.18 | 0.04 | 0 |
Clinical Oscillatory Instability Sepsis Detection
Overview
This dataset tests whether a model can detect oscillatory instability in a sepsis-like clinical system.
Many complex systems do not collapse immediately. Instead they begin to exhibit oscillatory behavior, where key system variables fluctuate in increasingly unstable cycles.
These oscillations often indicate that the system is approaching a critical boundary where small disturbances can trigger rapid deterioration.
The goal of this benchmark is to determine whether models can recognize early oscillatory instability before collapse occurs.
Prediction target
label_oscillatory_instability
0 = no oscillatory instability
1 = oscillatory instability present
The task is to detect oscillatory patterns that signal proximity to instability.
Row structure
Each row represents a synthetic clinical scenario.
Columns:
scenario_id
pressure
buffer_capacity
coupling_strength
oscillation_amplitude
oscillation_frequency
trajectory_drift
Training rows include the label.
Tester rows omit the label.
Evaluation
The scoring script reports:
accuracy
precision
recall
f1
specificity
negative predictive value (npv)
Primary metric
recall
Secondary metric
f1
Recall is prioritized because detecting oscillatory instability early can prevent system collapse.
Why this benchmark matters
Oscillatory behavior is a common precursor to collapse in complex systems.
In clinical settings oscillations in physiological signals can indicate that regulatory mechanisms are failing.
Detecting these oscillations allows earlier intervention and better stabilization strategies.
This benchmark tests whether models can detect temporal instability patterns rather than relying on static indicators.
Structural note
This dataset exposes system geometry while keeping the generator used to produce the scenarios private.
The goal is to evaluate whether models can detect oscillatory instability rather than memorizing patterns in static datasets.
Clarus Stability Geometry Benchmarks
This dataset is part of a broader benchmark family exploring instability and recovery in complex systems.
Related probes include:
clinical-compensation-collapse-sepsis-v1
clinical-fork-point-sepsis-transition-v1
clinical-organ-failure-cascade-v1
clinical-recovery-window-sepsis-v1
clinical-intervention-alignment-sepsis-v1
clinical-recovery-stability-sepsis-v1
clinical-false-stability-sepsis-v1
clinical-instability-margin-sepsis-v1
clinical-intervention-competition-sepsis-v1
Together these benchmarks map the lifecycle of instability and recovery in clinical dynamical systems.
License
MIT
- Downloads last month
- 16