Computer Aided Diagnosis System for Mammogram Abnormality

Detection of suspicious masses in mammograms play a vital role in early diagnosis of breast cancer, to reduce the death rate among women. The presence of masses and calcification’s in mammograms is distinguishing signs for breast cancer diagnosis. But in some cases, due to contrast variation, fuzzy boundaries and presence of noise in mammograms, segmentation and detection of masses is challenging assignment. This paper presents a new segmentation approach to detect masses in mammographic images. The proposed approach consist of artifacts elimination and pectoral region extraction, suspicious mass enhancement using dual morphological operation technique and finally, extraction of Regions of Interest (ROIs) from background using scaled Reyni entropy. The proposed system has been tested on two data-sets i.e. mini- Mammographic Image Analysis Society (mini-MIAS) and Digital Database for Screening Mammography (DDSM), over 50 and 90 mammograms respectively. Performance achieved with proposed system in terms of True Positive Fraction (TPF) yields 93.2% and 93.9% respectively, at the rate of 1.48 and 0.74 average False Positive per Image (FP/I), tested on both Cranio-Caudal (CC) and Medio-Lateral Oblique (MLO) views. The obtained experimental results demonstrates that proposed method gives improved results for mass detection and can be useful for radiologists in diagnosis of breast cancer at early stage.

Computer aided detection and classification system has been developed to detect breast cancer at an early stage by predicting the area and texture of malignant tumours. Noise removal and image enhancement is carried out in the pre-processing stage by using adaptive median filter and contrast limited histogram equalisation techniques. Improved watershed segmentation technique with appropriate internal and external markers, have proved to be an efficient approach in detecting the region of interest. The detected tumours are classified using feedforward artificial neural network that are trained using textural features. Area and entropy features extracted from malignant tumours aids in early detection of breast cancer by categorising malignant tumours as belonging to stage I or stage II. The overall efficiency of the system, for identifying stages of malignant tumour is 92%, which has been identified to be high when compared to all existing systems. Mammogram images from Mammographic Image Analysis Society (MIAS) database was used for training the system and efficiency of the system was tested using real time hospital images.

This study presents a computer aided system for automatic detection and classification of breast cancer in mammogram images. First the suspicious region or the Region of Interest is identified and extracted using Intuitionistic Fuzzy C-Means Clustering technique. Next multilevel Discrete Wavelet Transformation is applied to the extracted Region of Interest. After applying Discrete Wavelet Transformation, histogram features, Gray Level Concurrence wavelet features, and wavelet energy features are extracted from each Region of Interest of the image. Before classification, Principal Component Analysis is applied on the extracted features to reduce the feature dimension. Finally, the feature database is submitted to self-adaptive resource allocation network classifier for classification. The proposed system is verified with 295 mammograms in the Mammographic Image Analysis Society Database. The result shows that the proposed algorithm produces better results.

In this paper, we present a novel approach to detect microcalcifications (MCCs) in the mammograms. Microcalcification has approximately a size of several pixels in diameter on early cancer mammograms and has low contrast. So it is difficult to detection the microcalcifications. The essential idea of propose approach is to apply morphological and wavelet transform to locate the micro calcifications in digital mammograms. The presented method overcome higher true positive rate (TPR) but higher false positive (FP) problem by using single wavelet transform or morphology. The experiment results show the presented method has higher TPR and reducing FP obviously. Using the mammogram data form the Mammographic Image Analysis Society (MIAS) database the TPR is 80.2% at the cost of average 2.5 false microcalcifications per region of interest (ROI)

Strong evidence shows that characteristic patterns of breast tissues as seen on mammography, referred to as mammographic parenchymal patterns, provide crucial information about breast cancer risk. Quantitative evaluation of the characteristic mixture of breast tissues can be used as for mammographic risk assessment as well as for quantification of change of the relative proportion of different breast tissue patterns. This paper investigates mammographic segmentation based on spatial moments and prior information of mammographic building blocks (i.e. nodular, linear, homogenous, and radiolucent) as described by Taba¿r's tissue models to describe parenchymal patterns. The algorithm extracted texture features from a set of sub-sampled mammographic patches. Taba¿r's mammographic building blocks were modelled as statistical distribution of clustered filter responses based on spatial moments. Evaluation was based on the Mammographic Image Analysis Society (MIAS) database. The experimental results indicated that the developed methodology is capable of modelling complex mammographic images and can deal with intraclass variation and noise aspects. The results show realistic segmentation on tissue specific regions with respect to breast anatomy and Taba¿r's tissue models. In addition, the segmentation results were used for mammographic risk based classification of the entire MIAS database resulting in ~70% correct low/high risk classification.

This paper presents a machine learning based approach for the discrimination of malignant and benign microcalcification (MC) clusters in digital mammograms. A series of morphological operations was carried out to facilitate the feature extraction from segmented microcalcification. A combination of morphological, texture, and distribution features from individual MC components and MC clusters were extracted and a correlation-based feature selection technique was used. The clinical relevance of the selected features is discussed. The proposed method was evaluated using three different databases: Optimam Mammography Image Database (OMI-DB), Digital Database for Screening Mammography (DDSM), and Mammographic Image Analysis Society (MIAS) database. The best classification accuracy (%) was achieved for OPTIMAM using a stack generalization classifier with 10-fold cross validation obtaining an A value equal to .

A technique is proposed to classify regions of interests (ROIs) of digitized mammograms into mass and non-mass regions using texture features and artificial neural network (ANN). Fifty ROIs were extracted from the MIAS MiniMammographic Database, with 25 ROIs containing masses and 25 ROIs containing normal breast tissue only. Twelve texture features were derived from the gray level co-occurrence matrix (GLCM) of each region. The sequential forward selection technique was used to select four significant features from the twelve features. These significant features were used in the ANN to classify the ROI into either mass or non-mass region. By using leave-one-out method on the 50 images using the four significant features, classification accuracy of 86% was achieved for ANN. The test result using the four significant features is better than the full set of twelve features. The proposed method is compared with some existing works and promising results are obtained.

In the field of radiology, mammographic screened images (i.e. X-ray image sensing) are very challenging and difficult to interpret. The expert radiologist visually hunts the mammograms for any specific abnormality. However, human factor causes a low degree of precision that often results in biopsy and anxiety for the patient involved. This paper proposes a novel computer-aided detection (CAD) system to reduce the human factor involvement and to help the radiologist in automatic diagnosis of malignant/nonmalignant breast tissues by utilizing polar complex exponential transform (PCET) moments as texture descriptors. The input region of interest is extracted manually and subjected to further number of preprocessing stages. Both magnitude and phase of PCET moments are used for feature extraction of suspicious region. Moreover, a new classifier adaptive differential evolution wavelet neural network is introduced to improve the classification accuracy of the proposed CAD system. The proposed system is tested on the mammographic images from Mammographic Image Analysis Society database. The designed system attains a fair accuracy of 97.965% with 98.196% sensitivity and 97.194% specificity. The best area under the receiver operational characteristics curve for the proposed classifier is found to be 0.984 with confidence interval from 0.968 to 0.999 and ±0.0108 standard error.

Architecture distortion (AD) is an important and early sign of breast cancer, but due to its subtlety, it is often missed on the screening mammograms. The objective of this study is to create a quantitative approach for texture classification of AD based on various texture models, using support vector machine (SVM) classifier. The texture analysis has been done on the region of interest (ROI) selected from the original mammogram. A comprehensive analysis has been done on samples from three databases; out of which, two data sets are from the public domain, and the third data set is for clinical evaluation. The public domain databases are IRMA version of digital database for screening mammogram (DDSM) and Mammographic Image Analysis Society (MIAS). For clinical evaluation, the actual patient's database has been obtained from ACE Healthways, Diagnostic Centre Ludhiana, India. The significant finding of proposed study lies in appropriate selection of the size of ROIs. The experiments have been done on fixed size of ROIs as well as on the ground truth (variable size) ROIs. Best results pertain to an accuracy of 92.94% obtained in case of DDSM database for fixed-size ROIs. In case of MIAS database, an accuracy of 95.34% is achieved in AD versus non-AD (normal) cases for ground truth ROIs. Clinically, an accuracy of 88% was achieved for ACE dataset. The results obtained in the present study are encouraging, as optimal result has been achieved for the proposed study in comparison with other related work in the same area.

Alzheimer’s disease (AD) is an irreversible ailment. This ailment causes rapid loss of memory and behavioral changes. Recently, this disorder is very common among the elderly. Although there is no specific treatment for this disorder, its diagnosis aids in delaying the spread of the disease. Therefore, in the past few years, automatic recognition of AD using image processing techniques has achieved much attraction. In this research, we propose a novel framework for the classification of AD using magnetic resonance imaging (MRI) data. Initially, the image is filtered using 2D Adaptive Bilateral Filter (2D-ABF). The denoised image is then enhanced using Entropy-based Contrast Limited Adaptive Histogram Equalization (ECLAHE) algorithm. From enhanced data, the region of interest (ROI) is segmented using clustering and thresholding techniques. Clustering is performed using Enhanced Expectation Maximization (EEM) and thresholding is performed using Adaptive Histogram (AH) thresholding algorithm. From the ROI, Gray Level Co-Occurrence Matrix (GLCM) features are generated. GLCM is a feature that computes the occurrence of pixel pairs in specific spatial coordinates of an image. The dimension of these features is reduced using Principle Component Analysis (PCA). Finally, the obtained features are classified using classifiers. In this work, we have employed Logistic Regression (LR) for classification. The classification results were achieved with the accuracy of 96.92% from the confusion matrix to identify the Alzheimer’s Disease. The proposed framework was then evaluated using performance evaluation metrics like accuracy, sensitivity, F-score, precision and specificity that were arrived from the confusion matrix. Our study demonstrates that the proposed Alzheimer’s disease detection model outperforms other models proposed in the literature.

Breast cancer is alone accounted for 30% of all new cancer diagnoses for women. Early identification of breast cancer is a key stage in the process of diagnosis and treatment. Segmenting of low contrast mammograms is not an easy task due to irregular mass form, speculated margins and variation in intensity levels. Most of the region-based technique works well for homogenous images but fails for intensity inhomogeneity. Multiphase level set segmentation is optimised by the cuckoo search approach for energy minimisation. The proposed multiphase level set with the cuckoo optimisation method performs well in segmenting homogeneous as well as inhomogeneous regions. The Chan Vese (CV) and the proposed method are applied to the Mammographic Image Analysis Society (MIAS) database to obtain the accurate contour of the masses. To validate the proposed approach using quantitative performance measures such as Jaccard coefficient (JC), dice coefficient (DC), Hausdorff distance (Hd), sensitivity, specificity and accuracy were used for the MIAS data set. The results indicate that the suggested method can successfully segment the breast mass in mammograms and more robust than the classical CV method.

In this paper, an anatomic features-based method is proposed to segment the pectoral muscle region according to texture feature and shape feature. Firstly, spectral clustering combined with edge information is presented and used to segment the pectoral muscle, leading to an over-segmented result. Afterwards, a region merging algorithm is proposed on the basis of the intensity distribution of regions and the characteristic of triangle shape. Finally, the pectoral muscle is identified from the over-segmented result according to region merging. The proposed method is evaluated on 322 mammograms from the mammographic image analysis society (mini-MIAS) database and 50 mammograms from the Breast Center of Peking University People s Hospital (BCPKUPH). The results show that the proposed method works well for pectoral muscles of different sizes, shapes and intensities.

A proper and accurate classification technique with detection of brain tumor has been presented and proposed. The system uses neural network based approach for brain and breast image classification. Now a day's Magnetic resonance imaging (MRI technique is used for early detection of any abnormal changes in tissues and organs. The projected method is evaluated on two distinct databases i.e. Clinical database is database of brain MRI and one more Standard Digital Database for Screening Mammography (DDSM). The proposed system consists of Preprocessing using image normalization, morphological operations using erosion, dilation and Anisotropic Diffusion Filter (ADF), Extraction of texture feature using gray level co-occurrence matrix (GLCM) and classification into normal and abnormal using machine learning algorithm and quantization techniques i.e. LVQ. The proposed system achieved the accuracy of 68.85% for DDSM mammography database and 79.35% on clinical brain MRI database.

In this paper we will present a computer-aided detection (CAD) system for mass detection and classification. This CAD system performs mass detection on regions of interest (ROIs) and performs the normal-abnormal classification on detected masses. To enhance the ROIs, histogram equalization and average filtering were employed. ROIs or suspicious lesions were detected by an edge-base segmentation technique. Each ROI was represented by 12 textural features calculated from spatial gray-level co-occurrence matrix (GLCM). Finally, a back propagation neural network was used to classify a ROI to normal or abnormal one. Performance of the proposed system was analyzed in mini-MIAS data set by means of receiver operating characteristics curve (ROC curve) and free-response ROC curve (FROC curve). We archived area under the ROC curve AZ = 0,815. This value lied in the range from 0.7 to 0.9 so that our CAD system could be considered accurate enough.

Assessment of Neural Network Based Classifiers to Diagnose Focal Liver Lesions Using CT Images