Article reference:

M. Egmont-Petersen, E. Pelikan. "Detection of bone tumours in radiographs using neural networks," Pattern Analysis and
Applications, Vol. 2, No. 2, pp. 172-183, 1999.
 
 

Abstract:

We develop an approach for segmenting radiographic images of focal bone lesions possibly caused by bone tumour. A neural network is used to classify individual pixels by a convolution operation based on a feature vector. We design eight features chat characterise the local texture in the neighbourhood of a pixel. Four of the features are based on co occurrence matrices computed from the neighbourhood. The true class label of the pixels in the radiographs are obtained from annotations made by an experienced radiologist. We make a comparison of several statistical classifiers based on different criteria and argue that neural networks are most suited for our application. Feed-forward neural networks and self-organising feature maps are trained to perform the segmentation cask. The experiments confirm che feasibility of using a feature-based neural network for finding pathologic bone changes in radiographic images. An analysis of the eight features indicates that the presence of edges and transitions, the complexity of the texture, as well as the amount of high frequencies in the texture, are the main features discriminating (soft) tissue from pathologic bone, the two classes most likely to be confused.

Electronic reprint, or contact me: michael * egmont-petersen.nl (with * indicating @)