Early Detection: Machine Learning Techniques in Pancreatic Cancer Diagnosis
Mallipudi Devi Siva Sai
*
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
Palaparthi Prudhvi
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
Gollapudi M Naga Venkata Sai Gopi
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
Indla Ganeswara Naga Sai Ram
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
Mandadi Ram Sandeep
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
NagaBabu Pachhala
Department of Information Technology, Vasireddy Venkatadri Institute of Technology, Guntur, India.
*Author to whom correspondence should be addressed.
Abstract
Pancreatic cancer is a malignant tumor that poses a significant threat to patients' lives. Malignant growth is the abnormal development of cell tissue. Pancreatic illness is one of the most obvious causes of mortality across the world. Pancreatic malignant development begins in the pancreatic tissues. The pancreas secretes proteins that aid in digestion as well as hormones that direct sugar breakdown. Pancreatic cancer is typically identified in its late stages, spreads quickly, and has a terrible prognosis. Biomarkers are critical in the management of patients with invasive malignancies. Pancreatic Ductal Adenocarcinoma has a dismal prognosis due to its advanced appearance and limited treatment choices. This is compounded by the lack of validated screening and predicting biomarkers for early detection and precision therapy, respectively. In this paper, we have attempted to discuss various Machine Learning methods to detect pancreatic cancer. The selected. urinary biomarkers values are provided as the input of Support Vector Machine (SVM), Extra Tree Classifier (ETC), Decision Tree (DT), and Random Forest (RF) methods. The diagnosing accuracy of pancreatic cancer using SVM, ETC, DT, and RF classifiers are 50, 82.16, 81.03, and 86 respectively. The experimental results prove that the Random Forest classifier is more feasible and promising for clinical applications for the diagnosis of pancreatic cancer when compared to ETC, DT, and SVM.
Keywords: Early detection, machine learning, random forest algorithm, SVM, classification, data pre-processing, prediction