Mass detection in digital mammograms using gabor filter bank

Hussain, Muhammad; Khan, Salabat; Muhammad, Ghulam; Berbar, M.A.; Bebis, George

doi:10.1049/cp.2012.0465

Cited by 16 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During an initial training stage, a set of prototype features is computed at every texture pattern of interest (normal/tumor). The training images associated with each pattern are first filtered by applying a multichannel Gabor filter bank, obtaining a cloud of texture feature vectors for every pattern [5][6][7]. A set of prototypes is then extracted in order to represent that cloud.…”

Section: Proposed Methodologymentioning

confidence: 99%

See 1 more Smart Citation

Breast Masses Identification through Pixel-Based Texture Classification

et al. 2014

View full text Add to dashboard Cite

Abstract. Mammographic image analysis plays an important role in computer-aided breast cancer diagnosis. To improve the existing knowledge, this paper proposes a new efficient pixel-based methodology for tumor vs non-tumor classification. The proposed method firstly computes a Gabor feature pool from the mammogram. This feature set is calculated through multi-sized evaluation windows applied to the probabilistic distribution moments, in order to improve the accuracy of the whole system. To deal with a high dimensional data space and a large amount of features, we apply both a linear and non-linear pixel classification stage by using Support Vector Machines (SVMs). The randomness is encoded when training each SVM using randomly sample sets and, in consequence, randomly selected features from the whole feature bank obtained in the first stage. The proposed method has been validated using real mammographic images from well-known databases and its effectiveness is demonstrated in the experimental section.

show abstract

Section: Proposed Methodologymentioning

confidence: 99%

“…A two-dimensional Gabor filter defined as a Gaussian kernel modulated by an oriented complex sinusoidal wave can be described as follows [5,8,6]:…”

Section: Texture Feature Extractionmentioning

confidence: 99%

Breast Masses Identification through Pixel-Based Texture Classification

et al. 2014

View full text Add to dashboard Cite

show abstract

“…As a result, pixels on and near to an edge are detected by the SUSAN algorithm. The higher intensity distance a pixel has with an edge, the lower n(r 0 ) is set for it and the higher value is set as an output of function (3). In this work, since the output image of directional SUSAN algorithm is sufficient to find the breast boundaries, the edge enhancement method is not applied.…”

Section: Susan Edge Detectormentioning

confidence: 99%

Directional SUSAN image boundary detection of breast thermogram

Mahmoudzadeh

Zekri

Montazeri

et al. 2016

IET Image Processing

View full text Add to dashboard Cite

Thermography of the breast has been shown to be well suited to detect early signs of breast cancer. This study proposes a novel method for boundary detection of breast thermography images using directional SUSAN method. Among breast thermography image processing steps, breast isolation from background and from each other is an essential stage for proper detection of breast cancer. For this purpose, in this study, breast boundary is grouped into three regions depending on the region property. The algorithm of boundary detection is different for each region. Specially, for bottom breast boundary, directional SUSAN edge detector is presented that uses two rectangle masks to create a directional SUSAN gradient image with emphasis on oblique useful edges and omitting undesirable ones. Then cubic parabolic interpolation is implemented to determine a set of edge points on the boundaries. At last, an effective search algorithm is executed to correct some false points in order to extract breast boundaries accurately. The performance of the proposed approach illustrated by applying on the images of three databases. Experimental results show that this method acts effectively and confirm the accurate boundary detection. Moreover, statistical measures are calculated to indicate the remarkable capabilities of the proposed approach.

show abstract

“…Their method used textural features (Correlation, Energy, Entropy, Homogeneity and sum of Square variance) to train a neural network with 200 mammograms (100 normal, 100 cancer) and tested this with 50 mammograms (25 normal, 25 cancer). Mass classification using gabor filter features and SVM for classification is reported in [13].…”

Section: Literature Reviewmentioning

confidence: 99%

Hybrid Technique Based on N-GRAM and Neural Networks for Classification of Mammographic Images

Kulkarni

Stranieri

Kulkarni

et al. 2014

Computer Science &Amp; Information Technology ( CS &Amp; IT )

View full text Add to dashboard Cite

Various texture, shape, boundary features have been used previously to classify regions of interest in radiological mammograms into normal and abnormal categories. Although, bag-ofphrases or n-gram model has been effective in text representation for classification or retrieval of text, these approaches have not been widely explored for medical image processing. Our purpose is to represent regions of interest using an n-gram model, then deploy the n-gram features into a back-propagation trained neural network for classifying regions of interest into normal and abnormal categories. Experiments on the benchmark miniMIAS database show that

show abstract

Mass detection in digital mammograms using gabor filter bank

Cited by 16 publications

References 14 publications

Breast Masses Identification through Pixel-Based Texture Classification

Breast Masses Identification through Pixel-Based Texture Classification

Directional SUSAN image boundary detection of breast thermogram

Hybrid Technique Based on N-GRAM and Neural Networks for Classification of Mammographic Images

Contact Info

Product

Resources

About