No abstract

A method is developed to characterise highly irregular shape patterns, especially those appearing in biomedical settings. A collection of best-fitting ellipsoids is found using principal component analysis, and features are defined based on these ellipsoids in four different ways. The method is defined in a general setting, but is illustrated using two-dimensional images of dimorphic yeast exhibiting pseudohyphal growth, three-dimensional images of cancellous bone and three-dimensional images of marbling in beef. Classifiers successfully distinguishes between the yeast colonies with a mean classification accuracy of 0.843 (SD = 0.021), and between cancellous bone from rats in different experimental groups with a mean classification accuracy of 0.745 (SD = 0.024). A strong correlation (R 2 = 0.797) is found between marbling ratio and a shape feature. Key aspects of the method are that local shape patterns, including orientation, are learned automatically from the data, and the method applies to objects that irregular in shape to the point where landmark points cannot be identified between samples.

Pseudohyphal growth of the dimorphic yeast Saccharomyces cerevisiae is analysed using two-dimensional top-down binary images. The colony morphology is characterized using clustered shape primitives (CSPs), which are learned automatically from the data and thus do not require a list of predefined features or a priori knowledge of the shape. The power of CSPs is demonstrated through the classification of pseudohyphal yeast colonies known to produce different morphologies. The classifier categorizes the yeast colonies considered with an accuracy of 0.969 and standard deviation 0.041, demonstrating that CSPs capture differences in morphology, while CSPs are found to provide greater discriminatory power than spatial indices previously used to quantify pseudohyphal growth. The analysis demonstrates that CSPs provide a promising avenue for analysing morphology in high-throughput assays.

No abstract

No abstract