Accurate and Stable Feature Selection Powered by Iterative Backward Selection and Cumulative Ranking Score of Features

George, G. Victo Sudha; Raj, V. Cyril

doi:10.17485/ijst/2015/v8i11/71766

Cited by 5 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…that the pixel which has x value happened in a horizontal neighboring to a pixel which has y value through direction. For calculating the GLCM need to set some parameters such as [15]:…”

Section: Calculate the Co-occurrence Matrix Of Gray-level (Glcm)mentioning

confidence: 99%

Brain Cancer classification Based on Features and Artificial Neural Network

Abbadi¹,

Kadhim²

2017

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

MRI (Magnetic resonance Imaging) brain tumor images Classification is a difficult task due to the variance and complexity of tumors. This paper proposed techniques to classify the MR human brain images. The proposed classification technique consists of three stages, namely, pre-processing, feature extraction and selection, and classification. Features are extracted by using the gray level co-occurrence matrices and the gray level run length matrices (GLCM & GLRLM), 18 features were determined from the image, then selected the most important features that saved to the database. In the final stage, the classifier based on probabilistic neural network (PNN) have been used to classify MRI brain images, the proposed algorithm is trained with 50 images of (Sarcoma, Anaplastic Astrocytoma, Meningioma, and Benign) and tested with 65 images. The accuracy of this method was up to 98%. Figure 6: PNN Architecture& Training Parameters PNN classification evaluationThe evalution of the perofmance and accuracy of the proposed PNN is dpending on the following: 1. The first method is to test the network on the same sample features that trained on, then show the network how to classify them. Figure 7 shows that the network classifies each sample feature exactly to there class type.

show abstract

“…that the pixel which has x value happened in a horizontal neighboring to a pixel which has y value through direction. For calculating the GLCM need to set some parameters such as [15]:…”

Section: Calculate the Co-occurrence Matrix Of Gray-level (Glcm)mentioning

confidence: 99%

Brain Cancer classification Based on Features and Artificial Neural Network

Abbadi¹,

Kadhim²

2017

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

show abstract

“…Over the past two decades, many studies brought forward specific algorithms for quantifying whole-genome expression behavior ( 16 – 18 ), and other approaches as binary classification of normal versus tumor samples ( 19 – 23 ). Usually these approaches put their focus on the genome coding regions.…”

Section: Introductionmentioning

confidence: 99%

Quantification of read species behavior within whole genome sequencing of cancer genomes for the stratification and visualization of genomic variation

et al. 2016

View full text Add to dashboard Cite

The cancer genome is abnormal genome, and the ability to monitor its sequence had undergone a technological revolution. Yet prognosis and diagnosis remain an expert-based decision, with only limited abilities to provide machine-based decisions. We introduce a heterogeneity-based method for stratifying and visualizing whole-genome sequencing (WGS) reads. This method uses the heterogeneity within WGS reads to markedly reduce the dimensionality of next-generation sequencing data; it is available through the tool HiBS (Heterogeneity-Based Subclassification) that allows cancer sample classification. We validated HiBS using >200 WGS samples from nine different cancer types from The Cancer Genome Atlas (TCGA). With HiBS, we show progress with two WGS related issues: (i) differentiation between normal (NB) and tumor (TP) samples based solely on the information structure of their WGS data, and (ii) identification of specific regions of chromosomal amplification/deletion and their association with tumor stage. By comparing results to those obtained through available WGS analyses tools, we demonstrate some of the novelties obtained by the approach implemented in HiBS and also show nearly perfect normal/tumor classification, used to identify known and unknown chromosomal aberrations. Finally, the HiBS index has been associated with breast cancer tumor stage.

show abstract