Performance Evaluation of Different Supervised Machine Learning Algorithms in Predicting Linear Accelerator Multileaf Collimator Positioning’s Accuracy Problem

Radiation Oncology is one of the businesses that employs Machine Learning to automate quality assurance tests so errors and defects can be reduced, avoided, or eliminated as much possible during tumor therapy using a Linear Accelerator with MultiLeaf Collimator (Linac MLC). The majority applications have used supervised learning algorithms rather than unsupervised algorithms. However, in most cases, there clear bias deciding which machine algorithm use. And prediction findings may less accurate result this bias. As result, study, an evidence presented for novel application Logistic Regression technique predict Linac MLC positioning accuracy, achieved 98.68 percent accuracy robust consistent performance across several sets data. was obtained by comparing various (i.e. Regression, Decision Tree, Support Vector Machine, Random Forest, Naive Bayes, K-Nearest Neighbor) MLC's problem leaves' displacement datasets labelled results training test datasets. For each method, two parameters were evaluate performance: receiver operating characteristics curve. Based on evaluation, right selection sequence proposed order achieve near-optimal leaf problem. bias, well negative side effects ineffective preventive maintenance plan avoid solve causes inaccurate such motor fatigue stuck problems) could occurred successfully avoided.