Significance analysis of qualitative mammographic features, using linear classifiers, neural networks and support vector machines

Mavroforakis, Michael E.; Georgiou, Harris V.; Dimitropoulos, Nikos; Cavouras, D.; Theodoridis, Sergios

doi:10.1016/j.ejrad.2004.12.015

Cited by 43 publications

(34 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…1. In (a), morphological shape types of breast cancer are depicted (adapted from [3]). In (b), a tumor with a stellated boundary (light gray), perpendicular to the viewing plane (e.g.…”

Section: Methodsmentioning

confidence: 99%

The File-Card-Browser View for Breast DCE-MRI Data

Glaßer

Scheil

Preim

et al. 2011

Bildverarbeitung Für Die Medizin 2011

View full text Add to dashboard Cite

Abstract. Dynamic contrast-enhanced magnetic resonance imaging of the breast is acquired for the detection of breast cancer. To rate a tumor to be benign or malignant, radiologists evaluate the tumor's morphology and its enhancement kinetics. We present a new multi planar reformatting (MPR) view, the File-Card-Browser View, to improve and complete the standard axial slice-based evaluation. We tested our technique with a tumor set containing 20 cases and present first results.

show abstract

“…1. In (a), morphological shape types of breast cancer are depicted (adapted from [3]). In (b), a tumor with a stellated boundary (light gray), perpendicular to the viewing plane (e.g.…”

Section: Methodsmentioning

confidence: 99%

The File-Card-Browser View for Breast DCE-MRI Data

Glaßer

Scheil

Preim

et al. 2011

Bildverarbeitung Für Die Medizin 2011

View full text Add to dashboard Cite

show abstract

“…Most of the time, a second stage is necessary to reduce these features, because there are too numerous. In the third stage of the algorithm, the numerical descriptors are fed to classification algorithms, which are application-independent, such as Support Vector Machine [4,5,6], neural networks [2,6,7,8], k-nearest neighbors [9], etc. The classification algorithms will decide, depending on their entries, which is the class of the image.…”

Section: Classification Evaluationmentioning

confidence: 99%

Evaluation for uncertain image classification and segmentation

Martin

Laanaya

Arnold-Bos

2006

Pattern Recognition

View full text Add to dashboard Cite

Each year, numerous segmentation and classification algorithms are invented or reused to solve problems where machine vision is needed. Generally, the efficiency of these algorithms is compared against the results given by one or many human experts. However, in many situations, the location of the real boundaries of the object as well as their classes are not known with certainty by the human experts. Moreover, only one aspect of the segmentation and classification problem is generally evaluated. In our evaluation method, we take into account both the classification and segmentation results as well as the level of certainty given by the experts. As a concrete example of our method, we evaluate an automatic seabed characterization algorithm based on sonar images.

show abstract

“…2 Another classification technique that is widely used for the diagnosis of breast tumors are support vector machines which are known to be especially suited for dealing with classification data with a non-linear decision boundary. [3][4][5] One of the shortcomings is the black-box nature of the model, whereas in Bayesian networks statistical dependences and independences between features are represented. In this study we compare both classification methods and use two techniques, namely dimension reduction by principal component analysis (PCA) and normality transformation, to further improve the accuracy rate of the classifiers.…”

Section: Introductionmentioning

confidence: 99%

Classification of mammographic masses using support vector machines and Bayesian networks

et al. 2007

View full text Add to dashboard Cite

In this paper, we compare two state-of-the-art classification techniques characterizing masses as either benign or malign, using a dataset consisting of 271 cases (131 benign and 140 malign), containing both a MLO and CC view. For a number of suspected regions in a digitized mammogram, 12 out of 81 calculated image features have been selected for investigating the classification accuracy of support vector machines (SVMs) and Bayesian networks (BNs). Additional techniques for improving their performance were included in their comparison: the Manly transformation for achieving a normal distribution of image features and principal component analysis (PCA) for reducing our high-dimensional data. The performance of the classifiers were evaluated with Receiver Operating Characteristics (ROC) analysis. The classifiers were trained and tested using a k-fold cross-validation test method (k=10). It was found that the area under the ROC curve (A z ) of the BN increased significantly (p=0.0002) using the Manly transformation, from A z = 0.767 to A z = 0.795. The Manly transformation did not result in a significant change for SVMs. Also the difference between SVMs and BNs using the transformed dataset was not statistically significant (p=0.78). Applying PCA resulted in an improvement in classification accuracy of both BNs and SVMs, from A z = 0.767 to A z = 0.786, and A z = 0.793 to A z = 0.799, respectively. The difference in classification performance between BNs and SVMs after applying PCA was small and not statistically significant (p=0.11).

show abstract

Significance analysis of qualitative mammographic features, using linear classifiers, neural networks and support vector machines

Cited by 43 publications

References 13 publications

The File-Card-Browser View for Breast DCE-MRI Data

The File-Card-Browser View for Breast DCE-MRI Data

Evaluation for uncertain image classification and segmentation

Classification of mammographic masses using support vector machines and Bayesian networks

Contact Info

Product

Resources

About