Dublin Core
Title
Cancer Cells Detection Using Supervised Machine Learning
Abstract
Cancerous cells invade and destroy the healthy tissue of the body, including organs. It often begins in one part of the body before spreading uncontrollably to other areas of the organism. According to the World Health Organization (WHO), cancer is the cause of death worldwide - taking around 10 million lives yearly. The predominant cancers are colon, breast, lung, rectum, and prostate. Early detection is crucial to increase survival chances tremendously. Machine Learning (ML) tools have the potential to recognize key features in complex datasets enabling the classification of low and high risk patients. This research focuses on the use of supervised machine learning algorithms, such as Artificial Neural Networks (ANNs), Bayesian Networks (BNs),
Support Vector Machines (SVMs), and Decision Trees (DTs) for the development of predictive models, expected to result in effective and accurate decision-making based on available scientific experiments. The results of the study have showcased high accuracy rates (above 90 percent) on all applied models, with the highest accuracy scores using Feed Forward Neural Networks (approx. 97 percent). The use of machine learning methods can enhance the overall understanding of cancer progression, early detection, and treatment; however, thorough medical validation from professionals is essential for these methods to be adopted into routine clinical practice. The idea of adopting machine learning in the medical field is not to substitute human intelligence but to aid patients in receiving faster healthcare.
Support Vector Machines (SVMs), and Decision Trees (DTs) for the development of predictive models, expected to result in effective and accurate decision-making based on available scientific experiments. The results of the study have showcased high accuracy rates (above 90 percent) on all applied models, with the highest accuracy scores using Feed Forward Neural Networks (approx. 97 percent). The use of machine learning methods can enhance the overall understanding of cancer progression, early detection, and treatment; however, thorough medical validation from professionals is essential for these methods to be adopted into routine clinical practice. The idea of adopting machine learning in the medical field is not to substitute human intelligence but to aid patients in receiving faster healthcare.
Keywords
cancer cells, machine learning, predictive models