A survey on cancer prediction and detection with data analysis
- PDF / 1,370,635 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 17 Downloads / 195 Views
REVIEW ARTICLE
A survey on cancer prediction and detection with data analysis Arunava Sankar Nath1 · Aparajita Pal2 · Somnath Mukhopadhyay3 · Kartick Chandra Mondal1 Received: 15 June 2018 / Accepted: 1 August 2019 © Springer-Verlag London Ltd., part of Springer Nature 2019
Abstract World Health Organization reports cancer as a leading cause worldwide in mortality and morbidity. Accurate and early cancer risk assessment in average- to high-risk population is vital in controlling the cancer-related suffering and mortality. Advanced bioinformatics and data mining techniques along with computer-aided cancer prediction and risk assessment are used extensively to assist in identifying the high-risk population as well as individual cancer diagnosis and treatment. An early detection minimizes the risk of cancer spreading to secondary sites and ensures appropriate treatment at the onset of the malignancy. The scope of our survey was to review over 90 publications centered around works done in the area of data analysis studies in the field of cancer prediction and detection. The motivation was to accumulate and categorize knowledge on the usage of data analytics for cancer prediction and detection. The aim was to do a comparative study of few of the major analytical approaches in cancer data analysis and highlight their effectiveness. Keywords Survey · Cancer · Prediction · Detection · Data · Analysis
1 Introduction Every cell’s renewal, proliferation and death are strictly controlled by the cells genetic makeup. The instant this strict genetic control breaks down, mutation sets in and a clonal evolution starts which leads inexorably toward cancer. The rouge cell starts multiplying with mutated genetic makeup and proliferates indefinitely [1–3]. Cancer is a cell’s venal attempt to immortality which kills the rest of the body, and as the body dies, the rogue cell dies along with it. International Agency for Research on Cancer (IARC), a specialized cancer agency of the World Health Organization, estimates that 18.6 M cancer cases were registered [4] and about 9.6 M
B
Kartick Chandra Mondal [email protected] Arunava Sankar Nath [email protected] Aparajita Pal [email protected] Somnath Mukhopadhyay [email protected]
1
Department of Information Technology, Jadavpur University, Kolkata, India
2
Department of Biophysics, Bose Institute, Kolkata, India
3
Department of Computer Science and Engineering, Assam University, Silchar, Assam, India
cancer mortality was reported worldwide in 2018. The figures are expected to spike to 16.3 M in mortality worldwide with 29.5 M cancer cases registered by 2040. Predicting future cancer occurrence is an area of constant uncertainty. Cancer prediction models are crucial in assessing and identifying individuals at high risk in a given population [5,6]. Cancer predictive models aid in the design and planning of clinical trials, the evaluation of treatments and preventive interventions [7–9]. Accurate predictions help in allocating resources by health planners and in
Data Loading...
