Pancreatic Cancer Detection Using Machine Learning

Computer Engineering

This paper delivers into the significant research being conducted on medical health systems, which is driving the development of computing systems with the latest innovations. These advancements are facilitating more efficient implementations of medical systems, including the automatic identification of health-related disorders. Among the most crucial health research areas is the prediction of cancer, which can manifest in various forms and affect different parts of the body. Pancreatic cancer, one of the most common and currently considered incurable cancers, is a primary focus. Previous studies have identified a panel of three protein biomarkers (LYVE1, REG1A, and TFF1) in urine that can help detect resectable PDAC (pancreatic ductal adenocarcinoma). This study aims to improve this panel by replacing REG1A with REG1B, using data extracted into CSV format. Creatinine is a protein commonly used as an indicator of kidney function, LYVE1 may assist in tumor spread, REG1B is associated with pancreatic regeneration, and TFF1 is linked to urinary tract regeneration and repair. Effective treatment of pancreatic cancer is challenging once it is diagnosed. However, machine learning and neural networks are showing promise for accurate real-time segmentation of pancreatic images for early diagnosis. This research explores how to analyze pancreatic tumors using ensemble approaches in machine learning. Preliminary data suggest that the proposed technique enhances classifier performance for the early diagnosis of pancreatic cancer.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.