Longevity Recommendation for Root Canal Treatment

Endodontic treatment has a high success rate; however, it still fails in many patients. It is usually attributed to different clinical and non-clinical factors. Therefore, it is crucial to avoid or even significantly reduce the prevalence of the most common causes of root canal treatment failure. This paper makes an attempt to find the different factors that are responsible for root canal (RCT) failure by using machine learning techniques like SVM, NB classifier, and logistic regression. From the provided data of 332 instances, it determines the clinical and non-clinical aspects that lead to the identification of failing RC teeth. The findings also reveal that the LR model has the highest accuracy (91.87) compared to the other two algorithms. This system also helps in determining the relationship between these parameters and their impact on the longevity of the root canal treatment using a machine learning models. A longevity recommendation can help doctors improve their practice by pointing out areas where they may have fallen short.