000			02440nam a22001937a 4500
003			OSt
005			20220107122846.0
008			181227b xxu||||| |||| 00| 0 eng d
040			_cIIITMK
100			_aMalu G _915124
245			_aTexture based analysis and classification of lesions in medical images _cMalu G
300			_aPHD Thesis 2016
500			_aThe aim of this work is to develop an automated lesion detection and diagnosis system for breast Dynamic Contrast Enhanced - Magnetic Resonance Imaging (DCE-MRI). Early detection and diagnosis reduce the high death rate of patients due to breast cancer. Mammography and MRI are the popular medical imaging modalities that are used for the detection of malignant lesions in breast images; both have advantages and disadvantages. MRI is nowadays, shown to be a promising adjunctive tool for the malignancy detection. An automated system for an easy and accurate diagnosis of breast cancer is the need of the hour. In this study a number of algorithms, methods and techniques in image processing, mathematical, geometrical, and statistical equations, data mining, machine learning, and biological techniques were considered. Most appropriate techniques were used for the topic and were fortunate to conduct the experiments with real data obtained from Regional Cancer Centre, which was validated with the help of two eminent and experienced radiologists of the centre. The study was focused on two major factors that exhibited malignancy property - Enhancement and Structural. The enhancement property was studied using kinetic features and the structural property was performed using shape and margin characteristics of lesion. Since the aim of the work was to develop a fully automated lesion detection and diagnosis system, the steps involved in the development was bit extensive. Different techniques, experimentations and evaluations required to malignancy detection were used. It included pre-processing, image registration, segmentation; region of interest identification (ROI), developing new shape and margin descriptors, deriving features from the circular mesh based labelling technique, feature extraction, feature selection and classification.
502			_bPhD in Computer Science _dINT _eDr. Elizabeth Sherly IIITMK _cNovember 2016
650			_aMEDICAL IMAGE PROCESSING _915125
650			_aBREAST CANCER _915126
650			_aLESION DETECTION _915127
942			_2ddc _cTS
999			_c6292 _d6292