Bayesian Doctor

1.0.1

Stressed? Reason out the Root Cause.

Video Demonstration for Bayesian Doctor

Apply symptoms based probabilistic Reasoning in your Diagnostics

Realization of Baye's Theorem.

Bayes’ Theorem is a theorem of probability theory stated by the Reverend Thomas Bayes. The main idea of Baye's theorem is to update the probabilities of unobserved events based on observed evidence. Given a problem with unknown causes, when a new piece of evidence is observed, the prior beliefs of the unknown causes changes. Baye's theorem is used in a wide variety of contexts including Medicine, Marine Biology, Biomonitoring, Image processing, Spam Filter etc.

Bayesian Doctor is a tool for learning and applying diagnostics using Bayes’ theorem.

Elicit the Probability of a Diseases based on symptoms

Using the Bayesian Network directed graphical model, model the Symptoms and Causes as Random Variables, fill up the conditional dependency probabilities and get the hint of a possible disease.

Rich visual modeling of a Bayesian Network

By using a directed graphical model, Bayesian Network describes random variables and conditional dependencies. When you need to learn about the Bayesian network, there cannot be any excuse not to use an intuitive learning tool. Our Bayesian Doctor is the answer to that problem.

Create a Bayesian network in a very intuitive way, and find the Root cause hypotheses.

Bayesian_ Sprinkler_ Example

Query the Bayesian Network

Query_the_bayesian_network

Instantiate Observation

When you observe a state of a variable, instantiate the state of the Variable. For example, we observed that Grass is Wet, so we instantiated the Grass Variable = True. Based on the Observation, the Probability of Rain = True is updated.

Instantiate_observation

Multi-State Variables

Not only just Yes/No or True/False type boolean states, rather a variable can have more than 2 states. The Conditional Probability tables are easy to use and scaling of probabilities are easily done.

Multiple_ States_of_a_ Random_ Variable

Update your belief step by step using Diachronic Interpretation

When you have many random variables to work with, a Bayesian Network Conditional Probability Table can be very complicated and becomes fairly impractical in real life diagnostics. Therefore, you need to incorporate one observation at a time and update the beliefs accordingly. Here, the diachronic interpretation comes into play.

“Diachronic” means that something is happening over time; When we receive new data (new observation), the probability of our hypotheses changes. Diachronic Interpretation is a systematic way to update our beliefs as we get new data. In a lab-like setting, you can add as many mutually exclusive and non-exclusive hypotheses, add as many experiments and observation you want, and run a causal discovery session to find out the probability of your hypotheses. Diachronic interpretation tool can be used for various diagnostics from healthcare to daily problems.

Diagnose complicated diseases based on symptoms arrive:

An example of understanding the cause of Nausea / Vomiting. Initial Hypotheses were Peptic Ulcer and Viral Gastroenteritis. The diachronic interpretation can be used to update the belief on Peptic Ulcer and Viral Gastroenteritis based on symptoms.

nausea_updated

causal_discovery_fever

See the complete belief update history with rich information

tooltip_belief_history

Apply Diachronic Interpretation even to find the root cause of daily life issues.

Diachronic_ Interpretation_ Main

Supported Environments

Operating System

Windows 7, Windows 8-8.1, Windows 10

Microsoft .NET Framework

4.5 or later