scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

Automatic detection of tumor subtype in mammograms based On GLCM and DWT features using SVM

TL;DR: The proposed work increases the accuracy of classification and reduces the percentage of false positives in mammography images, since they are most effective, low cost and one of the highly sensitive techniques.
Abstract: Mammography images are employed in diagnosing breast cancers, since they are most effective, low cost and one of the highly sensitive techniques such that they can detect even small lesions. The proposed work increases the accuracy of classification and reduces the percentage of false positives. The images from the data set are initially preprocessed and contrast enhanced which makes the image most effective for further analysis. Then Region Of Interest (ROI) is determined from morphological top hat filtered image by means of thresholding segmentation. Various features like first order textural features, Gray Level Co-occurrence Matrix (GLCM) features, Discrete Wavelet Transform (DWT) features, run length features and higher order gradient features are derived for the particular ROI. Support Vector Machine (SVM) classifier is trained with the above mentioned features using MATLAB bioinformatics tool box. Thus the classified results are obtained for the query image based on the trained SVM structure. The mammography data set has been taken from the Mammographic Image Analysis Society (MIAS) in which there are 322 images available along with ground tooth information.
Citations
More filters
Journal ArticleDOI
TL;DR: A robust classification model for automated diagnosis of the breast tumor with reduction of false assumptions in medical informatics is presented and it is observed that rate of false positives decreased by the proposed method to improve the performance of classification, efficiently.
Abstract: Early screening of skeptical masses or breast carcinomas in mammograms is supposed to decline the mortality rate among women. This amount can be decreased more on development of the computer-aided diagnosis with reduction of false suppositions in medical informatics. Our aim is to provide a robust tumor detection system for accurate classification of breast masses using normal, abnormal, benign, or malignant classes. The breast carcinomas are classified on the basis of observed classes. This is highly dependent on feature extraction process. In propose work, a novel algorithm for classification based on the combination of top Hat transformation and gray level co-occurrence matrix with back propagation neural network. The aim of this study is to present a robust classification model for automated diagnosis of the breast tumor with reduction of false assumptions in medical informatics. The proposed method is verified on two datasets MIAS and DDSM. It is observed that rate of false positives decreased by the proposed method to improve the performance of classification, efficiently.

64 citations

Proceedings ArticleDOI
01 Oct 2016
TL;DR: A novel and new coarse-to-fine method is proposed to segment the brain tumor using a hierarchical framework that consists of preprocessing, deep learning network based classification and post-processing.
Abstract: Accurate tumor segmentation is an essential and crucial step for computer-aided brain tumor diagnosis and surgical planning. Subjective segmentations are widely adopted in clinical diagnosis and treating, but they are neither accurate nor reliable. An automatical and objective system for brain tumor segmentation is strongly expected. But they are still facing some challenges such as lower segmentation accuracy, demanding a priori knowledge or requiring the human intervention. In this paper, a novel and new coarse-to-fine method is proposed to segment the brain tumor. This hierarchical framework consists of preprocessing, deep learning network based classification and post-processing. The preprocessing is used to extract image patches for each MR image and obtains the gray level sequences of image patches as the input of the deep learning network. The deep learning network based classification is implemented by a stacked auto-encoder network to extract the high level abstract feature from the input, and utilizes the extracted feature to classify image patches. After mapping the classification result to a binary image, the post-processing is implemented by a morphological filter to get the final segmentation result. In order to evaluate the proposed method, the experiment was applied to segment the brain tumor for the real patient dataset. The final performance shows that the proposed brain tumor segmentation method is more accurate and efficient.

50 citations

08 Feb 2017
TL;DR: In classification process, this research attempt to compare k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) classifier in order to achieve the better accuracy, and shows that SVM outperforms KNN in breast cancer abnormalities classification with 93.88% accuracy.
Abstract: In order to begin the initial check on breast cancer, radiologist can use Computer Aided Diagnosis (CAD) as another option to detect breast cancer. During breast cancer check, human error is often to affecting the result. Several research before have proved that CAD is able to detect breast cancer spot more accurate. The purpose of this research is to find reliable method to classify breast cancer abnormalities. Mammography Image Analysis Society (MIAS) database is used as the sample data to the proposed system in this research. Mammograms are divided into three categorize which are normal, benign and malignant according to MIAS database. Features included in this experiment are extracted by using gray level co-occurrence matrices (GLCM) at 0o, 45o, 90o and 135o with a block size of 128x128. In classification process, this research attempt to compare k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) classifier in order to achieve the better accuracy. The result shows that SVM outperforms KNN in breast cancer abnormalities classification with 93.88% accuracy.

20 citations


Additional excerpts

  • ...[6] presented the first order and gradient features combined with GLCM, Discrete Wavelet Transform (DWT), run length and higher order gradient features to detect the breast cancer in mammogram images....

    [...]

Journal ArticleDOI
01 Feb 2016-Optik
TL;DR: A system for the detection of calcifications, based on a new approach suggested of pretreatment of image mammography, is proposed, which allows extracting successfully the calcifications starting from the mammography referents from the mini-MIAS database.

11 citations

Journal ArticleDOI
TL;DR: A new algorithm to detect the pattern of a microcalcification by calculating its physical characteristics is proposed, which acts as a good classifier that is used to classify the considered input image as normal or abnormal with the help of only two physical characteristics.
Abstract: Breast cancer is one of the life-threatening cancers occurring in women. In recent years, from the surveys provided by various medical organizations, it has become clear that the mortality rate of females is increasing owing to the late detection of breast cancer. Therefore, an automated algorithm is needed to identify the early occurrence of microcalcification, which would assist radiologists and physicians in reducing the false predictions via image processing techniques. In this work, we propose a new algorithm to detect the pattern of a microcalcification by calculating its physical characteristics. The considered physical characteristics are the reflection coefficient and mass density of the binned digital mammogram image. The calculation of physical characteristics doubly confirms the presence of malignant microcalcification. Subsequently, by interpolating the physical characteristics via thresholding and mapping techniques, a three-dimensional (3D) projection of the region of interest (RoI) is obtained in terms of the distance in millimeter. The size of a microcalcification is determined using this 3D-projected view. This algorithm is verified with 100 abnormal mammogram images showing microcalcification and 10 normal mammogram images. In addition to the size calculation, the proposed algorithm acts as a good classifier that is used to classify the considered input image as normal or abnormal with the help of only two physical characteristics. This proposed algorithm exhibits a classification accuracy of 99%.

10 citations

References
More filters
Journal ArticleDOI
01 Mar 2001
TL;DR: An easy-to-use intelligent system that gives the user options to diagnose, detect, enlarge, zoom and measure distances of areas in digital mammograms and finds that a combination of three features is the best combination to distinguish a benign microcalcification pattern from one that is malignant.
Abstract: An intelligent computer-aided diagnosis system can be very helpful for radiologist in detecting and diagnosing microcalcification patterns earlier and faster than typical screening programs. In this paper, we present a system based on fuzzy-neural and feature extraction techniques for detecting and diagnosing microcalcifications' patterns in digital mammograms. We have investigated and analyzed a number of feature extraction techniques and found that a combination of three features (such as entropy, standard deviation and number of pixels) is the best combination to distinguish a benign microcalcification pattern from one that is malignant. A fuzzy technique in conjunction with three features was used to detect a microcalcification pattern and a neural network was used to classify it into benign/malignant. The system was developed on a Microsoft Windows platform. It is an easy-to-use intelligent system that gives the user options to diagnose, detect, enlarge, zoom and measure distances of areas in digital mammograms.

219 citations


"Automatic detection of tumor subtyp..." refers methods in this paper

  • ...The Back Propagation Neural Network (BPNN) classifier used in [5] achieved classification rate of 83....

    [...]

Journal ArticleDOI
TL;DR: The results demonstrate the feasibility of the authors' approach to the development of a CAD system for DBT mammography and demonstrate the ability of this system to be applied to breast mass detection on digital breast tomosynthesis mammograms.
Abstract: The purpose of the study was to design a computer-aided detection (CAD) system for breast mass detection on digital breast tomosynthesis (DBT) mammograms and to perform a preliminary evaluation of the performance of this system. Twenty-six patients were imaged with a prototype DBT system. Institutional review board approval and written informed patient consent were obtained. Use of the data set in this study was HIPAA compliant. The CAD system first screened the three-dimensional volume of the mass candidates by means of gradient-field analysis. Each mass candidate was segmented from the structured background, and its image features were extracted. A feature classifier was designed to differentiate true masses from normal tissues. The CAD system was trained and tested by using a leave-one-case-out method. The classifier calculated a mean area under the test receiver operating characteristic curve of 0.91 ± 0.03 (standard error of mean). The CAD system achieved a sensitivity of 85%, with 2.2 false-positive...

129 citations

Journal ArticleDOI
01 Nov 2008
TL;DR: In this paper, the texture properties of the tissue surrounding micro calcification (MC) clusters on mammograms for breast cancer diagnosis were investigated using a probabilistic neural network, which achieved an area under receiver operating characteristic curve (Az) of 0.989.
Abstract: The current study investigates texture properties of the tissue surrounding microcalcification (MC) clusters on mammograms for breast cancer diagnosis. The case sample analyzed consists of 85 dense mammographic images, originating from the digital database for screening mammography. mammograms analyzed contain 100 subtle MC clusters (46 benign and 54 malignant). The tissue surrounding MCs is defined on original and wavelet decomposed images, based on a redundant discrete wavelet transform. Gray-level texture and wavelet coefficient texture features at three decomposition levels are extracted from surrounding tissue regions of interest (ST-ROIs). Specifically, gray-level first-order statistics, gray-level cooccurrence matrices features, and Lawspsila texture energy measures are extracted from original image ST-ROIs. Wavelet coefficient first-order statistics and wavelet coefficient cooccurrence matrices features are extracted from subimages ST-ROIs. The ability of each feature set in differentiating malignant from benign tissue is investigated using a probabilistic neural network. Classification outputs of most discriminating feature sets are combined using a majority voting rule. The proposed combined scheme achieved an area under receiver operating characteristic curve (Az) of 0.989. Results suggest that MCspsila ST texture analysis can contribute to computer-aided diagnosis of breast cancer.

127 citations

Journal ArticleDOI
16 Oct 2004
TL;DR: An algorithm for detecting masses in mammographic images acquired in several hospitals belonging to the MAGIC-5 collaboration by means of a ROI Hunter algorithm, without loss of meaningful information is presented.
Abstract: The mammography is the most effective procedure for an early diagnosis of the breast cancer. In this paper, an algorithm for detecting masses in mammographic images will be presented. The database consists of 3762 digital images acquired in several hospitals belonging to the MAGIC-5 collaboration (Medical Applications on a Grid Infrastructure Connection). A reduction of the whole image's area under investigation is achieved through a segmentation process, by means of a ROI Hunter algorithm, without loss of meaningful information. In the following classification step, feature extraction plays a fundamental role: some features give geometrical information, other ones provide shape parameters. Once the features are computed for each ROI, they are used as inputs to a supervised neural network with momentum. The output neuron provides the probability that the ROI is pathological or not. Results are provided in terms of ROC and FROC curves: the area under the ROC curve was found to be AZ=0.862plusmn0.007, and we get a 2.8 FP/Image at a sensitivity of 82%. This software is included in the CAD station actually working in the hospitals belonging to the MAGIC-5 Collaboration

124 citations


"Automatic detection of tumor subtyp..." refers methods in this paper

  • ...al and using thresholding in [4] by Casico....

    [...]

Proceedings Article
23 Jul 2002
TL;DR: This paper illustrates, by comparison to other published research, how important the data cleaning phase is in building an accurate data mining architecture for image classification.
Abstract: This paper proposes a new classification method based on association rule mining. This association rule-based classifier is experimented on a real dataset; a database of medical images. The system we propose consists of: a preprocessing phase, a phase for mining the resulted transactional database, and a final phase to organize the resulted association rules in a classification model. The experimental results show that the method performs well reaching over 80% in accuracy. Moreover, this paper illustrates, by comparison to other published research, how important the data cleaning phase is in building an accurate data mining architecture for image classification.

124 citations