Prediction of Bevel-Tip Needle Deflection Using CART Based Decision Tree

Minimal-invasive surgery has gained popularity due to reduced blood loss, minimal patient discomfort, lower costs, and faster recovery. However, its success relies heavily on the needle tip's precision. The interaction between the needle and tissue notably influences this accuracy. Achieving a significant degree of accuracy is challenging because of the uncertain mechanics and the varying stiffness of heterogeneous tissue. The foremost factor influencing the misplacement of a needle is deflection. Therefore, this research paper introduces a machine learning approach for predicting needle deflection. The machine learning method such as a decision tree, is utilized and trained using collected experimental data. Decisions, such as determining which feature contributes more to deflection, are made without depending on any mathematical model.