Objective: Recognition of orofacial disharmonies and offering appropriate treatment plans are intelligent actions which are usually performed by an orthodontist. Can we help orthodontists in diagnosis and treatment planning using a computerized algorithm? That is the question we are trying to address using intelligent agents.
Methods: Using object oriented theory in the computer science; we can model any intelligent agent in an object that has its own "properties and methods". We are trying to find a solution model using the necessary knowledge in the form of "properties of an object". On the other hand, by means of artificial intelligence algorithms, we will make the methods of mentioned object to form "intelligent behavior" for it.
Results: The result would be an intelligent agent that takes its required data from different sources like patient, processed images or the dentist. Based on gathered data, the intelligent agent processes the information internally to deliver its findings either as a diagnosis or treatment plan; or as a clue for orthodontist to do more investigations.
Conclusions: To show how we can use Artificial Neural Networks to offer a treatment plan more clearly, we used object oriented methodology and created a base design and documented our design using class diagrams. As a sample, we discussed an orthodontic case and used a very simple linear neural network to offer treatment plan. We also showed how the system may learn during the time to produce more effective plans.