Application of Artificial Intelligence in Tumors Sizing Classification for Breast Cancer.
Ricardo González-Otal, José Luis López Guerra, Carlos Luís Parra-Calderón, Alicia Martínez-García, Alberto Moreno-Conde, María José Ortiz-Gordillo · IWBBIO · 2013
The staging in breast cancer is one of the most important prognostic factors. However, the complex coding TNM criteria, which includes clinical and pathological components, the existence of different versions of TNM classification guides over time, and the variability of the source used to obtain data, makes the manual collection of TNM staging in free text be variable and imprecise. The aim of this project is to develop a tool based on artificial intelligence that allows the collection of tumor size (T) staging data for breast cancer automatically, reducing the variability. Our approach, based on two steps, starts with the detection and extraction of tumor's size characteristics in free text, using a simple natural language processor. Secondly, based on the data extracted, we applied different data mining algorithms for the T classification such as the J48 classifier tree, LADtree and NaiveBayes. Then, structured TNM reports for patients are created.