Evaluation of Machine Learning Methods for Pancreatic Cancer Detection Using CT Scans

The application of machine learning algorithms in medical image processing has emerged as a possible means of achieving this goal. Early identification of pancreatic cancer is essential for enhancing patient outcomes. In this study, we studied how to identify pancreatic cancer utilising CT scan images using Convolutional Neural Networks (CNN), K-Nearest Neighbours (KNN), Support Vector Machines (SVM), and Random Forest (RF). To thoroughly test these algorithms, a dataset of 2000 CT scan images was meticulously divided into training and testing subsets. With an accuracy rate of 94.34 %, the CNN model performed remarkably well, establishing a new standard for accuracy in the diagnosis of pancreatic cancer. In addition to exhibiting outstanding accuracy rates above 88 %, KNN, SVM, and RF also shown their potential as therapeutic instruments. Furthermore, we thoroughly evaluated the prediction skills of the models using performance metrics like precision, recall, F1 score, and accuracy. This study promises prompt intervention and better patient outcomes, which is a significant advance in the early diagnosis of pancreatic cancer. These findings inspire further investigation and innovation in the area of medical image processing in addition to highlighting the transformational potential of machine learning in medical diagnostics.