Using Support Vector Machines (Svm) to Facilitate Classification of Cancer

Abstract

Scientists are now able to concurrently perform screening or expression of many genes using DNA micro-arrays to establish their status that is whether they are silent, hyperactive, or active in cancerous or normal tissue [1]. Given the fact that such new micro-array devices produce massive volumes of critical data, there is a need to develop new analytical methods that are capable of sorting out whether the cancer cell shave unique signatures of gene tissue over the several types of cancer tissue or the normal cells. The current multiple data sets which are publicly available on open source platforms such as Internet are characterized by many problems, for instance,a rather smaller number of experiments and a huge amount of gene expression values per experiment. Additionally, it is evident that data analysis can be conducted from diverse perspectives. The majority of the empirical studies in the available relate to gene clusters which have been revealed through unsupervised learning methodologies. Clustering is usually conducted together with other data dimensions; for instance, each experiment might link to one patient who might carry or not have a certain disease. Therefore, clustering is mostly done with the intention of grouping the patients with comparable clinical history or record. Presently, it is evident that the application of supervised learning to classify cancer and proteins has increasingly gained prominence.

Authors and Affiliations

Ray Nick

Keywords

Related Articles

Reversal with Sugammadex in the Absence of Neuromuscular Monitoring

Sugammadex provides rapid and reliable recovery of neuromuscular (NM) function from rocuronium or vecuronium-induced NMblockade than neostigmine. The dose of sugammadex, ranges from 2 to 16 mg/kg, is adequately matched w...

Factor VII Deficiency and Sepsis in Pregnancy Treated with Recombinant Factor VIIa

Inherited factor VII deficiency is a rare autosomal recessive coagulation disorder, classified as type 1 or type 2 depending on the absence or presence of factor VII antigen in plasma. There are only few published case r...

SMOFlipid versus Intralipid in Postoperative ICU Patients

Aim of the work Lipids are important components of total parentral nutrition, especially for patients after major abdominal surgery. Traditionally used intralipid has many complications and can lead to increased infectio...

Does Preoperative Carbohydrate Intake Reduces Postoperative Stress Response in Patients Undergoing Valve Replacement Surgeries

Background Cardiac surgery is a stress that causes insulin resistance leading to increased insulin requirements. The aim of our study was to evaluate whether preoperative oral intake of carbohydrate rich drinks could imp...

Patient Controlled Epidural Analgesia (PCEA) with or without Background Infusion using Fentanyl and Bupivacaine for Major Upper Abdominal Surgery

Background Pain thresholds vary in individuals. Need for analgesia thereby differs in individuals. Methods The aim was to compare, patient controlled epidural analgesia (PCEA) with demand bolus (Group A) versus demand b...

Download PDF file
  • EP ID EP626177
  • DOI -
  • Views 168
  • Downloads 0

How To Cite

Ray Nick (2018). Using Support Vector Machines (Svm) to Facilitate Classification of Cancer. Enliven: Journal of Anesthesiology and Critical Care Medicine, 5(3), 6-8. https://europub.co.uk/articles/-A-626177