Medical Diagnosis Simulator

Saudi Healthcare Scenario

You're a physician at King Faisal Specialist Hospital in Riyadh. A patient presents with symptoms, and you need to determine the most likely diagnosis using Bayesian reasoning. This simulator demonstrates how medical diagnosis works using probabilistic inference.

How Bayesian Diagnosis Works

1. Priors: Base rates of diseases in the population
2. Likelihoods: Probability of symptoms given each disease
3. Evidence: Symptoms the patient actually has
4. Posteriors: Updated probabilities after seeing evidence

Patient Profile

Age: 35 years

Location: Riyadh, Saudi Arabia

Occupation: Office worker

Step 1: Select Patient Symptoms

Check all symptoms the patient is experiencing:

Fever (≥38°C)
Elevated body temperature

Cough
Dry or productive cough

Fatigue
Extreme tiredness

Headache
Head pain or migraine

Sore Throat
Throat pain or irritation

Loss of Taste/Smell
Anosmia or ageusia

Bayesian Calculation Progress

Prior Probabilities

Likelihoods

Evidence Calculation

Posterior Results

Step 1: Prior Probabilities (Base Rates)

These are the background probabilities of each disease in the Saudi population:

Disease	Prior Probability	Interpretation

Note: These priors are based on typical prevalence rates in Saudi Arabia during respiratory season.

Step 2: Symptom Likelihoods P(Symptom|Disease)

For each disease, what's the probability of observing each symptom?

Disease	Fever	Cough	Fatigue	Headache	Sore Throat	Taste Loss

Understanding Likelihoods

These values come from medical literature and clinical studies. For example, COVID-19 has a high likelihood of causing loss of taste/smell (anosmia), while bacterial infections are more likely to cause high fever.

Saudi Healthcare Scenario

How Bayesian Diagnosis Works

Patient Profile

Step 1: Select Patient Symptoms

Bayesian Calculation Progress

Step 1: Prior Probabilities (Base Rates)

Step 2: Symptom Likelihoods P(Symptom|Disease)

Understanding Likelihoods

Step 3: Evidence Calculation

Raw Scores (Likelihood × Prior):

Normalization (Divide by Total):

Step 4: Final Diagnosis Results

Posterior Probabilities