10. Literature Review: Overview of Cancer Treatment and Prediction Approaches Based on Machine Learning

The purpose of this chapter is to provide a literature survey through an overview of the research fields relevant to cancer treatment and prediction approaches based on machine learning. The past few years have witnessed an exponential growth in databases and repositories due to the increase in scientific knowledge and the massive data production. Biomedical domain represents one of the rich data domains. An extensive amount of biomedical data is currently available with wealth of information, ranging from details of clinical symptoms to various types of biochemical data and outputs of imaging devices El Houby (J Appl Biomed, 2018 [1]). One of the important biomedical research domains, epidemiological cancer research, who is of high priority across the world. Cancer has been characterized as a heterogeneous disease consisting of many different subtypes. Every sixth death in the world is due to cancer, making it the second-leading cause of death (second only to cardiovascular diseases) Schutte (Global, regional, and national age-sex specific mortality for 264 causes of death, 1980–2016: a systematic analysis for the Global Burden of Disease Study, 2017 [2]). The early diagnosis and prognosis of a cancer type have become a necessity in cancer research, as it can facilitate the subsequent clinical management of patients. The vast amount of hidden data in huge databases related to cancer with all his variability has created tremendous interests in the field of data mining. While data mining is a discipline resulting from the combination of classical statistics and computer science algorithms, such as machine learning, aim to the extraction of new and useful knowledge from a large amount of data, it has become a useful instrument in bioinformatics. It can depict variation of cancer incidence and mortality by region, ethnicity, gender and socioeconomic factors that contribute to the assessment of population health needs, while it can contribute to the study of cancer burden. Furthermore, in-depth analysis of the patient’s profile using data mining methods may uncover hidden, previously unknown relations between patient profile, cancer treatment and surveillance. This chapter will present recent approaches and works related to this context. It presents also a synthesis of current and future trends for smart systems for E-health.