Significant patterns extraction to find most effective treatment for oral cancer using data mining

Abstract

Development of cancer in oral mucosa as classified by the World Health Organization is a two-stage process that initially shows up as a premalignant, precancerous sore and that subsequently develops into the malignant cancerous stage. Early evaluation of oral precancerous lesions has a dramatic impact on oral cancer mortality rates as the medicine is very effective in early stage diagnosis. This paper aims at extracting the patterns that help finding the most effective course of oral cancer treatment and its post-treatment management. The Apriori algorithm is used to mine a set of significant rules for prevention of oral cancer by adopting the most efficient treatment. We attempt to find the association among various treatments, histopathology, follow-up symptoms, and follow-up examination. The experimental results show that all the generated rules hold the highest confidence level, thereby making them very useful for deciding effective treatment to cure oral cancer and its follow-up.