A Two Phase Hybrid Classifier based on Structure Similarities and Textural Features for Accurate Meningioma Classification

Kiran Fatima,Hammad Majeed

doi:10.14569/ijacsa.2017.080460

Abstract

Meningioma subtype classification is a complex pattern classification problem of digital pathology due to het-erogeneity issues of tumor texture, low inter-class and high intra-class texture variations of tumor samples, and architec-tural variations of cellular components. The basic aim is the achievement of significantly high classification results for all the subtypes of meningioma while dealing with inherent complexity and texture variations. The ultimate goal is to mimic the prognosis decision of expert pathologists and assist newer pathologists in making right and quick decisions. In this paper, a novel hybrid classification framework based on nuclei shape matching and texture analysis is proposed for classification of four subtypes of grade-I benign meningioma. Meningothelial and fibroblastic subtypes are classified on basis of nuclei shape matching through skeletons and shock graphs while an optimized texture-based evolutionary framework is designed for the classification of transi-tional and psammomatous subtypes. Classifier-based evolutionary feature selection is performed using Genetic Algorithm (GA) in combination with Support Vector Machine (SVM) to select the optimal combination of higher-order statistical features extracted from morphologically processed RGB color channel images. The proposed hybrid classifier employed leave-one-patient-out 5-fold cross validation and achieved an overall 95.63% mean classification accuracy.

Highlights

Now-a-days, Computer-Aided Diagnosis (CAD) is in wide practice in clinical work for the detection and prognosis of various types of abnormalities
Grey-level cooccurrence matrix (GLCM) features extracted from the most stable subbands are classified through Support Vector Machine (SVM) classifier with 88% accuracy. 5-fold cross validation is performed by using one patient data for testing and data of remaining four patients from each subtype for training
The histology images of meningioma subtypes shown in Fig. 1. have a clear visual representation of nuclei shapes owned by different subtypes

Summary

Introduction

Now-a-days, Computer-Aided Diagnosis (CAD) is in wide practice in clinical work for the detection and prognosis of various types of abnormalities. CAD has emerged as one of the key research areas in medical imaging, diagnostic pathology and radiology. The variability in the opinion of different pathologists and the variation in the opinion of a single pathologist at different timings may be observed in the prognosis decision of a tumor specimen [1]. Due to these reasons, adaptive approaches and quantitative methods for computer-aided diagnosis are needed to assist histopathologists for correct diagnosis and prognosis of tumors

Methods

Findings

Discussion

Conclusion