A review on current and future prospective of Cancer Classification through Deep Learning
2019
Srihariakash K | Poornima R M | Prakash Balu | Gayathri. | V | Yashicka J V | Shen-Ming Chen
Cancer is the second foremost origin of death in the world, next to heart disease. The name cancer refers to more than a thousand sicknesses illustrate by out of direct development & replication of multiple cells. Due to this reason of cancer analysis, utilization of microarray datasets along with machine learning methods escalating in the current research scenario. Classification is one of the very broadly used datamining techniques to build a model that describes & distinguishes data classes in a manner to be used to predict the class of unseen instances. In machine learning, features are chosen manually for a classifier. With Deep learning features, extraction and modelling steps are automatic. Deep learning is one of the most significant among machine learning that requires computing system to iteratively perform calculations to identified patterns by itself. Deep learning use training data to discover underlying patterns, build models & make predictions based on the best fit model. In the last decades, there has been a growing interest of addressing cancer classification using deep learning due to their positive revival of neural networks and connectionism from the genuine integration of the latest advances in parallel processing enabled by coprocessors. Here the review of deep learning for classification in bioinformatics presenting examples of current research. Additionally, we discuss Deep learning and convolutional neural network working principles to provide a useful and comprehensive perspective, this paper presents three works DeepGen, SDAE, Enhance Feature learning in a brief description of each study. We believe that this review will provide valuable insights and serve as a starting point for the researcher to apply deep learning approaches for classification in Gene expression dataset
显示更多 [+] 显示较少 [-]