Internet-based system for diagnosis of coronary artery disease
Jeffrey J. Lovelace, Krzysztof J. Cios, Dorel M. Sala, Lucy S. Goodenday · 2002
An intelligent Internet-based coronary artery disease (CAD) diagnostic system based on Tl-201 planar imaging data was developed. The system can self-generate rules to correctly diagnose obstructions in the three main arteries, namely right coronary artery (RCA), left anterior descending (LAD), circumflex (CCX), or a patient being normal (with no obstructed arteries). The rule formation engine is based on the CLIP3 (Cover Learning using Integer Programming version 3) machine learning algorithm. The system stores all data (training data, fuzzy sets and rules) into a relational database. The primary consideration of the authors' design was to keep the system easy to use, thus the database operations are transparent to the user. The architecture of the system is a multithreaded structure that allows multiple operations to be carried out in parallel. The biggest advantage of the system is that a diagnosis can be carried out over the Internet via HTML forms.