Self-Learning Model-Based Segmentation of Medical Images

Interaction increases flexibility of segmentation but it leads to undesirable behaviour of an algorithm if knowledge being requested is inappropriate. In region growing, this is the case for defining the homogeneity criterion, as its specification depends also on image formation properties that are not known to the user. We developed a region growing algorithm that learns its homogeneity criterion automatically from characteristics of the region to be segmented. The method is based on a model that describes homogeneity and simple shape properties of the region. Parameters of the homogeneity criterion are estimated from sample locations in the region. These locations are selected sequentially in a random walk starting at the seed point, and the homogeneity criterion is updated continuously. In contrast to other adaptive region growing methods our approach produces results that are far less sensitive to the seed point location, and it allows a segmentation of individual structures. The model-based adaptive region growing approach was extended to a fully automatic and complete segmentation method by using the pixels with the smallest gradient length in the not yet segmented image region as a seed point. Both methods were tested for segmentation on test images and of structures in CT images. The performance of the semi-automatic method is compared with the adaptive moving mean value region growing method, and the automatic method is compared with the watershed segmentation. We found our method to work reliable if the model assumption on homogeneity and region characteristics were true. Furthermore, the model is simple but robust thus allowing for a certain amount of deviation from model constraints and still delivering the expected segmentation result.