Data structure and algorithms for a medical diagnosis software

Question

For an academic question, I plan to design a medical diagnosis system. Given a description of symptoms, produce a list of probable diseases (with closeness matching). I'm having some trouble finding papers for getting started. Which algorithms and data structure should I consider?

Answer 1

Machine learning algorithms and techniques are very successful and best practice in this area (clustering and classification techniques could be uses). For example in Machine Learning Techniques for AD/MCI Diagnosis and Prognosis or Medical decision support systems based on machine learning, thesis written by Chih-Lin Chi.

You can use simple records in database as the data model.

Answer 2

The field of probabilistic graphical models is often used for medical diagnosis, and some tools have been implemented. See this example, for a start.

Also see this.