Clinical Entropy & Counterfactual Harm Simulator v1.0-D3

Bayesian diagnostic uncertainty reduction & iatrogenic risk balancing

Initial Entropy
2.12 bits
5 Prior Candidates
Current Entropy
0.84 bits
-60.38% uncertainty
Top Probability
0.88
Acute Coronary Syn.
Harm Index
0.080
Optimal Risk Window

Differential Diagnosis State Space Network

Node Size = Probability | Node Link = Diagnostic Path
Entropy Decay Curve (bits) Sequence Step: 2 / 2 ordered

Order Diagnostic Procedures

Toggle tests to observe Bayesian updates

Bayesian Posterior Distribution

Counterfactual Harm Analysis

Missed Rx Risk vs. Invasive Risk
Expected Missed Diagnosis Harm: 0.050
Iatrogenic Procedure Harm: 0.030
Combined Net Clinical Harm: 0.080

Trace the saved watermark counting rules and fixture limits

Read the explanation

The saved clinical simulator computes minus the sum of probability times log base two probability. An illustrative uniform two-option distribution has entropy one bit; four uniform options have two bits. Both bars use three hundred pixels per bit. Entropy measures uncertainty within this entered model. These hypothetical distributions explain its algebra and do not establish a diagnosis, calibrated clinical probabilities or a treatment recommendation. For each selected test the code assumes a positive result, multiplies the targeted option by sensitivity and others by one minus specificity, then renormalizes. In an illustrative two-option case, prior one half times sensitivity point eight gives target weight point four. The other prior one half times false-positive probability point two gives point one. Both bars show unnormalized probability weight at one thousand pixels per unit. Normalization gives point eight and point two. These are illustrative inputs, not clinical test performance evidence. Its assigned missed-harm contribution is probability times one minus probability times an authored harm weight. For illustrative weight one, probability one half yields point two five, while probability point nine yields point zero nine. Both bars show assigned contributions at two thousand pixels per unit. Procedure harm is then summed separately. This pedagogical score is not a measured patient outcome or validated clinical risk assessment.

Super generates helpful tools and automates fact-checking across the internet proactively. If you enjoyed this tool, build your own with Super and share it with a friend.