Classification in to non overlapped groups of Z

Classification of  pulmonary nodules using Novel Z with Tilted Z
Local Binary Pattern (Z?TZLBP)

 

Abstract

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

 

In
this paper , a novel feature vector named Z with Tilted Z Local Binary Pattern (Z?TZLBP)
is proposed for extracting pulmonary nodule image features powerfully. The goal
is to reduce LBP’s complexity by reducing the size of the feature vector. In
the proposed work by dividing the vicinity pixels in to non overlapped groups
of Z and TZ(Tilted Z) further texture features of pulmonary images are
extracted for feature extraction. The Single Classifier KNN has been used with
different distance metrices for classification purpose. Metric Accuracy and
F-measure is used to evaluate the performance of  the proposed system.  

 

Keywords    Classification , Pulmonary Image , Feature
Extraction , Active Contour , KNN

 

1 Introduction

 

Pulmonary images are very much
important to detect lung diseases using CT imaging modality for the assessment
of pulmonary nodules. The feature types of the pulmonary nodule in CT images
are important cues for the malignancy prediction 1-2, diagnosis and advance
management 3-4.The texture features of nodule solidity and semantic
morphology feature of speculation are critical to differentiate of pulmonary
nodules and other subtypes. Meanwhile other semantic features calcification
pattern, roundedness, margin clearness are shown to be helpful for the
evaluation of nodule classification. The nodule may be found in bronchial tubes
or outside of bronchial tube. If the nodule<=3mm the detection of malignancy is difficult. The determination of clinical characteristics may differ from patient to patient and also depends on experience of the observer. Computer aided diagnosis is an assistive software package to provide computational diagnostic  references for the clinical image reading and decision making support. The histogram features for the high level texture analysis helped to extract nodule feature. Ciompi et al5 developed the bag of frequencies descriptor that can successfully distinguished 51 spiculated nodules from the other 204 non-spiculated nodules. However the mapping from the lowlevel image features toward the high level semantic features in the domain of clinical terms is not straight forward task. This semantic feature assessment maybe useful for clinical analysis. The Lung Image Database Consortium (LIDC) dataset for its rich annotation database supports the training and testing CAD scheme6-7. The nodules  diameters  larger than  3mm  are  further  rated  by radiologist referred with semantic features of  "subtlety", "calcification", "spericity", "margin", "speculation", "texture",  "lobulation",  "internal structure" and "malignancy"8. Absorption and scattering of light rays are the two major issues that cause reduced quality of images. Several methods have been proposed to enhance the quality of the pulmonary images. Histogram equalization technique , Contrast stretching methods capable to enhance the image quality. Contrast Limited Adaptive Histogram Equlization (CLAHE) has been applied to improve the image contrast. Otsu's adaptive thresholding method form image segmentation has been effective for many applications.  This provides bright backgrounds for images. Various thresholding techniques such as Local, Global and Multilevel thesholding have been applied for the segmentation of  pulmonary nodules images.The texture feature descriptor that has been widely used for image classification is Local Binary Pattern. Pican et al. have used GLCM's  twenty four types features for extraction and for each image suitable features have to be chosen for extraction. Hence there is a need of  efficient feature descriptor for classification process. In past years Neural Network is performed  for classification results time consuming. K-Nearest Neighbor as classifier with Euclidean distance used to classify nodules. Padmavathi et al15 have classified images using probalisitic neural network which gives better results than SIFT algorithm with three classes of dataset.  Eduardo et al have classified images using nine machine learning algorithms such as: Decision Trees, Random Forest, Extremely  Randomised Trees , Boosting, Gradient Boosted Trees, Normal Bayes Classifier,Expectation Maximization, NN and SVM. Bhuvaneswari.P et al16 have classified coral and textures using KNN by considering K=1 and the accuracy reported as 90.35%.