Application of Machine Learning and Deep Learning in Pancreatic Cancer Diagnosis: A Review

Pancreatic cancer remains one of the deadliest types of cancer due to late and poor diagnosis. While many artificial intelligence-based techniques have been applied to diagnose this type of cancer, no attempt has been made to review machine learning (ML), deep learning (DL), and ensemble learning methods on this type of cancer. This study addresses the gap by providing a comprehensive review and comparative analysis of artificial intelligence models specifically designed for pancreatic cancer diagnosis, and by combining findings across multiple studies. Our goal is to evaluate and compare various AI methods for diagnosing pancreatic cancer and identify the most promising approaches that can be utilized to enhance early-stage diagnosis. This review of studies from PubMed, Elsevier, Google Scholar, and MDPI databases was conducted to examine the application of the most commonly used algorithms, including ML, DL, and ensemble learning, for pancreatic cancer diagnosis. The accuracy levels of these algorithms were analyzed and compared. Results show AI algorithms, especially ensemble models, can enhance the accuracy of this cancer diagnosis and lead to improved quality of life and patient outcomes by personalizing treatments.

کلیدواژه ها:

Artificial intelligence ، Machine learning ، Deep learning ، Pancreatic cancer diagnosis