A Review on Analyzing and Predicting the State of Cancer Disease using Machine Learning Algorithms

Cancer represents a chronic anomaly induced by mutations within genes that regulate the functioning of body cells, particularly their growth and replication. Over 100 variations of cancer exist, each named after the tissue or organs in which they originate. Cancer stands as the leading global cause of death. Timely detection and prognosis of cancer have become imperative, aiding subsequent medical intervention for patients. This research offers a comparison of machine learning, deep learning, and optimization strategies for cancer detection. Numerous researchers have focused on cancer diagnosis and prognosis, with each approach exhibiting distinct accuracy rates influenced by specific circumstances, tools, and datasets. Navigating challenges associated with obtaining high-quality labeled datasets, addressing the diverse nature of cancer, and ensuring the ethical integration of machine learning into clinical workflows are essential obstacles. Additionally, dealing with class imbalance, extracting features from intricate biological data, ensuring dependable validation, and integrating various data sources pose crucial challenges for developing robust algorithms in cancer analysis. The main aim of this study is to evaluate various existing Machine Learning and optimization approaches, identifying the most suitable methods to accommodate extensive datasets with high prediction accuracy. This revision primarily aims to showcase previous research on machine-learning techniques employed for cancer detection. The study presents comprehensive information for novices wishing to explore machine learning methods and establish a foundation in deep learning.