https://csie.ntu.edu.tw/~rfchang/prof/ar0302.pdf

Support Vector Machines for Diagnosis of Breast Tumors on US Images

Background
Mammography has established itself as the most efficient technique for the identification of the pathological breast lesions. Among the various types of lesions, microcalcifications are the most difficult to identify since they are quite small (0.1-1.0 mm) and often poorly contrasted against an images background. Within this context, the Computer Aided Detection (CAD) systems could turn out to be very useful in breast cancer control.

/pdf/fuzzy-technique-for-microcalcifications-clustering-in-5dozm039of.pdf

Fuzzy technique for microcalcifications clustering in digital mammograms

Segmentation via morphological granulometric features
is based on fitting structuring elements into image topography from
below and above. Each structuring element captures a specific texture
content. This paper applies granulometric segmentation to digitized
mammograms in an unsupervised framework. Granulometries
based on a number of flat and nonflat structuring elements are computed,
local size distributions are tabulated at each pixel,
granulometric-moment features are derived from these size distributions
to produce a feature vector at each pixel, the Karhunen–Loeve
transform is applied for feature reduction, and Voronoi-based clustering
is performed on the reduced Karhunen–Loeve feature set.
Various algorithmic choices are considered, including window size
and shape, number of clusters, and type of structuring elements.
The algorithm is applied using only granulometric texture features,
using gray-scale intensity along with the texture features, and on a
compressed mammogram. Segmentation results are clinically
evaluated to determine the algorithm structure that best accords to
an expert radiologist’s view of a set of mammograms.

Unsupervised morphological granulometric texture segmentation of digital mammograms

A computationally efficient algorithm for computing erosions and dilations by one- dimensional grayscale structuring elements with constant slope is proposed. The computational complexity of this algorithm is independent of the size of the support of the structuring function. This is a generalization of the method proposed by Van Herk for the case of erosion and dilation by flat one-dimensional structuring elements. By appropriate combinations of these structuring elements, it is possible to approximate many useful structuring elements. This enables efficient computation of granulometries where the number of operations depends linearly on the number of openings. Theoretical and experimental results comparing the complexity of this algorithm with other standard techniques is presented. Two memory efficient algorithms are then presented. Several implementation issues in computing a granulometry and moments of the associated morphological pattern spectrum are then addressed. An efficient implementation of granulometries for large images on machines with limited memory, by dividing the image into smaller rectangular patches is then discussed. The optimum size of these patches is a function of the specific hardware and has been obtained experimentally for three different hardware platforms. Finally, parallel implementation of the different algorithms on two multi-processor machines is discussed.

A Constant-time Algorithm for Erosions/Dilations with Applications to Morphological Texture Feature Computation

Presents a comprehensive discussion on the segmentation of mammograms using morphological texture features. These features are derived from morphological granulometries with various structuring elements. Each structuring element captures a specific texture content. The segmentation is carried out in an unsupervised manner by applying the KL (Karhunen-Loeve) transform feature reduction and Voronoi clustering on the extracted morphological texture features. The evaluation of the segmentation outcome by a trained radiologist is provided.

Nasser Kehtarnavaz

Papers

Interactive Selective and Adaptive Clustering for Detection of Microcalcifications in Mammograms