Breast Density Segmentation: A Comparison of Clustering and Region Based Techniques

Original scientific paperMammography as an x-ray method usually gives good results for lower density breasts while higher breast tissue densities significantly reduce the overall detection sensitivity and can lead to false negative results. In automatic detection algorithms knowledge about breast density can be useful for setting an appropriate decision threshold in order to produce more accurate detection. Because the overall intensity of mammograms is not directly correlated with the breast density we have decided to observe breast density as a texture classification problem. In this paper we propose breast density classification using feature selection process for different classifiers based on grayscale features of first and second order. In feature selection process different selection methods were used and obtained results show the improvement on overall classification by choosing the appropriate method and classifier. The classification accuracy has been tested on the mini-MIAS database and KBD-FER digital mammography database with different number of categories for each database. Obtained accuracy stretches between 97.2 % and 76.4 % for different number of categories.Key words: Breast Density, Feature Selection, Haralick Features, Soh Features, Classification Klasifikacija dojki prema gustoći izborom značajki. Mamografija je rendgenska metoda koja daje dobre rezultate pri slikanju dojki koje imaju manju gustoću, dok joj osjetljivost značajno opada pri snimanju dojki veće gustoće i time može doći do lažno pozitivnih rezultata. Poznavanje gustoće dojke može biti korisno kod algoritama za automatsku detekciju zbog mogućnosti odredivanja praga odluke na osnovi tog znanja. S obzirom na to da ukupni intenzitet pojedinog mamograma nije izravno povezan s gustoćom, odlučili smo se promatrati gustoću kao problem klasifikacije teksture. U ovom radu predlažemo klasifikaciju dojki prema gustoći izborom izdvojenih značajki intenziteta prvog i drugog reda za različite klasifikatore. Za odredivanje prikladnih značajki koristili smo različite metode i tako dobivene značajke pokazale su bolju točnost klasifikacije za odabrane klasifikatore. Točnost klasifikacije testirali smo na bazi mamografskih slika mini-MIAS i bazi digitalnih mamografskih slika KBD-FER s različitim brojem kategorija u koje su slike bile podijeljene. Postignuta točnost klasifikacije proteže se izmedu 97,2 % i 76,4 % za različit broj kategorija u koje su mamogrami podijeljeni.

show abstract

“…Torrent et al presented a comparison of different approaches for clustering of fatty and dense breast tissue [14].…”

Section: Overview Of the Previous Workmentioning

confidence: 99%

Breast Density Classification Using Multiple Feature Selection

2012

View full text Add to dashboard Cite

show abstract

“…The results of the algorithm demonstrated close agreement to radiologist's segmentation and density interpretation. Torrent et al [19] compared two clustering based algorithm and one region based algorithm to segment fatty and dense tissue in mammographic images. The first algorithm is a multiple thresholding algorithm based on the excess entropy (EE), the second one is based on the Fuzzy C-Means clustering algorithm (FCM), and the third one Fisherfaces (FF) is based on a statistical analysis of the breast.…”

Section: Introductionmentioning

confidence: 99%

Density based breast segmentation for mammograms using graph cut techniques

Saidin

Ngah

Sakim

et al. 2009

TENCON 2009 - 2009 IEEE Region 10 Conference

View full text Add to dashboard Cite

Abstract-In this work we explore the application of graph cuts techniques to the problem of finding the boundary of different breast tissue regions in mammograms. The goal of the segmentation algorithm is to see if graph cuts algorithm could separate different densities for the different breast patterns. The graph cut is applied with multi-selection of seeds label to provide the hard constraint, whereas the seeds labels are selected based on user defined. Graph cuts have been explored on images of various imaging modalities but not on mammograms just yet. Therefore, this project is mainly focused on using graph cut algorithm to perform segmentation to increase visibility of different breast densities in mammography images. Segmentation of the mammogram into different mammographic densities is useful for risk assessment and quantitative evaluation of density changes. Our proposed methodology for the segmentation of mammograms on the basis of their region into different densities based categories has been tested on MIAS database.

show abstract

“…While some model-driven supervised approaches exist [2,8], the majority of segmentation algorithms utilize semisupervised or unsupervised approaches, particularly region growing [8,9], contour detection [8,10], polygon fitting [10], and image clustering [1,2,3,8]. Our work falls into the clustering category and represents the first application of WaveCluster to the mammographic domain.…”

Section: Mammographic Segmentationmentioning

confidence: 99%

“…Similar segmentation approaches used in the literature include fuzzy c-means [2,3] and watershed segmentation [5]. These techniques were compared in accuracy and runtime, using the following experimental setup:…”

Section: Comparisonmentioning

confidence: 99%

“…However, existing breast and ROI segmentation approaches, such as approaches based on k-means [1], fuzzy c-means [2,3], gray level co-occurrence matrices (GLCM) [4], and the watershed transform [5], suffer several shortcomings. For example, k-means and fuzzy c-means tend to find spherical clusters, k-means cannot differentiate between significant and nonsignificant clusters (i.e., all pixels receive a label), GLCM makes assumptions on the texture of the underlying region, and watershed transforms have high sensitivity to noise and a well-known tendency to oversegment (See Figure 1).…”

Section: Open Accessmentioning

confidence: 99%

See 1 more Smart Citation

Mammographic Segmentation Using WaveCluster

Barnathan¹

2012

Algorithms

View full text Add to dashboard Cite

Segmentation of clinically relevant regions from potentially noisy images represents a significant challenge in the field of mammography. We propose novel approaches based on the WaveCluster clustering algorithm for segmenting both the breast profile in the presence of significant acquisition noise and segmenting regions of interest (ROIs) within the breast. Using prior manual segmentations performed by domain experts as ground truth data, we apply our method to 150 film mammograms with significant acquisition noise from the University of South Florida's Digital Database for Screening Mammography. We then apply a similar segmentation procedure to detect the position and extent of suspicious regions of interest. Our approach was able to segment the breast profile from all 150 images, leaving minor residual noise adjacent to the breast in three. Performance on ROI extraction was also excellent, with 81% sensitivity and 0.96 false positives per image when measured against manually segmented ground truth ROIs. When not utilizing image morphology, our approach ran in linear time with the input size. These results highlight the potential of WaveCluster as a useful addition to the mammographic segmentation repertoire.

show abstract

Breast Density Segmentation: A Comparison of Clustering and Region Based Techniques

Cited by 22 publications

References 16 publications

Breast Density Classification Using Multiple Feature Selection

Breast Density Classification Using Multiple Feature Selection

Density based breast segmentation for mammograms using graph cut techniques

Mammographic Segmentation Using WaveCluster

Contact Info

Product

Resources

About