Boundary modelling and shape analysis methods for classification of mammographic masses

Abstract

The problem of computer-aided classification of benign and malignant breast masses using shape features is addressed. The aim of the study is to look at the exceptions in shapes of masses such as circumscribed malignant tumours and spiculated benign masses which are difficult to classify correctly using common shape analysis methods. The proposed methods of shape analysis treat the object's boundary in terms of local details. The boundaries of masses analysed using the proposed methods were manually drawn on mammographic images by an expert radiologist (JELD). A boundary segmentation method is used to separate major portions of the boundary and to label them as concave or convex segments. To analyse the shape information localised in each segment, features are computed through an iterative procedure for polygonal modelling of the mass boundaries. Features are based on the concavity fraction of a mass boundary and the degree of narrowness of spicules as characterised by a spiculation index. Two features comprising spiculation index (SI) and fractional concavity (fcc) developed in the present study when used in combination with the global shape feature of compactness resulted in a benign/malignant classification accuracy of 82%, with an area (Az) of 0.79 under the receiver operating characteristics (ROC) curve with a database of the boundaries of 28 benign masses and 26 malignant tumours. SI alone resulted in a classification accuracy of 80% with Az of 0.82. The combination of all the three features achieved 91% accuracy of circumscribed versus spiculated classification of masses based on shape.

Classification of breast masses using a committee machine of artificial neural networks

“…1͑d͔͒. 8,9,13,24 However, some benign masses can have lobulated or spiculated contours and blurred margins, and some maglignant tumors can have contours that are round or oval, as well as well-defined margins ͑see Fig. 1͒.…”

“…1͒. 8,9,13,24 To represent the features of breast masses as described before, several researchers have been studying image processing techniques to extract measures of shape, texture, and edge sharpness. 9-11,13,14,24 Quantitative features as described before have provided significant results in the discrimination between malignant tumors and benign masses in pattern classification studies.…”

“…9-11,13,14,24 Quantitative features as described before have provided significant results in the discrimination between malignant tumors and benign masses in pattern classification studies. 11,19,[22][23][24] In this work, we use the features proposed by Rangayyan et al 9,13 and Alto, Rangayyan, and Desautels, 24 which are described next.…”

“…The feature F cc is computed by taking the cumulative length of the concave segments and dividing it by the total length of the contour. 13 Benign masses ͓Fig. 2͑a͔͒ have fewer, if any, concave segments than malignant tumors ͓Fig.…”

“…With the objective to detect and extract the characteristics of lesions, several techniques of image processing have been proposed. [8][9][10][11][12][13][14][15] The artificial neural network ͑ANN͒ is an example of computational intelligence techniques that has been used to classify tumors related to breast cancer. 12,[16][17][18][19][20][21][22][23] Several types of network architecture, such as the multilayer perceptron ͑MLP͒, the single-layer perceptron ͑SLP͒, 19 and ra-dial basis functions ͑RBFs͒ 20 have been used for the classification of breast masses and tumors based on measures of shape, texture, and edge sharpness.…”