Handbook of Biomedical Image Analysis, Vol.2: Segmentation by Jasjit S. Suri (Editor), David Wilson (Editor), Swamy
By Jasjit S. Suri (Editor), David Wilson (Editor), Swamy Laxminarayan (Editor)
Read or Download Handbook of Biomedical Image Analysis, Vol.2: Segmentation Models Part B PDF
Best biophysics books
Additional info for Handbook of Biomedical Image Analysis, Vol.2: Segmentation Models Part B
Then the presence of pathological tissues, which are not included in the statistical model, is considered (d). Finally, a downsampling step introduces partial voluming in the model (e). Model-Based Brain Tissue Classiﬁcation 13 is illustrated in Fig. 5(c), which shows a sample of the total resulting image model. The MRF brings general spatial and anatomical constraints into account during the classiﬁcation, facilitating discrimination between tissue types with similar intensities such as brain and nonbrain tissues.
Although not always visible for a human observer, it can cause serious misclassiﬁcations when intensity-based segmentation techniques are used. 2 the mixture model is therefore extended by explicitly including a parametric model for the bias ﬁeld. 5(b) shows a typical sample of the resulting model. The model parameters are then iteratively estimated by interleaving three steps: classiﬁcation of the voxels; estimation of the normal distributions; and estimation of the bias ﬁeld. The algorithm is initialized with information from a digital brain atlas about the a priori expected location of tissue classes.
As a result, the total iterative scheme now consists in four steps, shown in Fig. 11. 11: The extension of the model with a MRF prior results in a four-step algorithm that interleaves classiﬁcation, estimation of the normal distributions, bias ﬁeld correction, and estimation of the MRF parameters. 24 Leemput et al. The calculation of the MRF parameters poses a difﬁcult problem for which a heuristic, noniterative approach is used. For each neighborhood conﬁguration (N p , N o ), the number of times that the central voxel belongs to class k in the current classiﬁcation is compared to the number of times it belongs to class k , for every couple of classes (k, k ).