A non-invasive methodology for the grade identification of astrocytoma using image processing and artificial intelligence techniques

作者：

Highlights：

• To suggest a non-invasive method for classification of Astrocytoma through rigorous training and testing.

• To compare and conclude the best medical image segmentation technique.

• To develop an efficient automatic feature selection technique for grade identification.

• To analyze and quantify the performance of classifiers constructed for the grade identification of Astrocytoma (tumor).

摘要

•To suggest a non-invasive method for classification of Astrocytoma through rigorous training and testing.•To compare and conclude the best medical image segmentation technique.•To develop an efficient automatic feature selection technique for grade identification.•To analyze and quantify the performance of classifiers constructed for the grade identification of Astrocytoma (tumor).

论文关键词：Astrocytoma,Image segmentation,Tumor isolation,Feature extraction,Feature selection,Grade classification,MR,Magnetic Resonance,PD,Proton Density,CSF,Cerebro Spinal Fluid,CT,Computer Tomography,FCM,Fuzzy Clustering Means,SVM,Support Vector Machines,LVQ,Learning Vector Quantization,PSO,Particle Swarm Optimization,GLCM,Gray Level Co-occurrence Matrix,ROI,Region Of Interest,SFLA,Shuffling Frog Leaping Algorithm,PCA,Principle Component Analysis,MRI,Magnetic Resonance Image,MRS,Magnetic Resonant Spectroscopy,PCNN,Pulse Coupled Neural Network

论文评审过程：Received 9 May 2014, Revised 19 August 2015, Accepted 21 August 2015, Available online 28 August 2015, Version of Record 20 October 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.08.036