Modeling of myocardial contractility using parameterized super-quadric SPECT images.

We developed methods to represent cardiac motility. Using an innovative model, we estimated several parameters of cardiac features. We implemented the parameterized super quadric model to visualize the motion of a left ventricle (LV) with OpenGL and Visual C++. We displayed myocardial wall thickening with a super-ellipsoidal model. The time frames in this model changed the measured thickening count. We also parameterized motility using the parameterized super quadric model. We analyzed the motility of the LV myocardium and tested its criteria using a validation study of seven normal subjects and 26 patients with prior myocardial infarction. To analyze motility, we used mean and variance of total motion during a cardiac cycle. The average of a normal subject was 0.46 and variance was 0.02. For patients, average and variance of motility were 0.59 and 0.08 respectively. Although the average value did not differ between normal subjects and patients, the variance differed significantly. Thus, we were able to estimate the difference between normal subjects and patients. In patients, motility was 128% higher than in normal subjects, and the variance was 328% higher. In the patient study, quantity of motion decreased rapidly in a stressed state. The visualization for contractility displayed 15 segment variables; we were able to rotate the locations of all points with a mouse interface. We were able to visualize most of the factors for cardiac motility and cardiac features. We expect that this model can distinguish between normal subjects and abnormal subjects, and that we can produce an exact analysis of momentum using this model.

Multi-resolution wavelet-transformed image analysis of histological sections of breast carcinomas.

Multi-resolution images of histological sections of breast cancer tissue were analyzed using texture features of Haar- and Daubechies transform wavelets. Tissue samples analyzed were from ductal regions of the breast and included benign ductal hyperplasia, ductal carcinoma in situ (DCIS), and invasive ductal carcinoma (CA). To assess the correlation between computerized image analysis and visual analysis by a pathologist, we created a two-step classification system based on feature extraction and classification. In the feature extraction step, we extracted texture features from wavelet-transformed images at 10x magnification. In the classification step, we applied two types of classifiers to the extracted features, namely a statistics-based multivariate (discriminant) analysis and a neural network. Using features from second-level Haar transform wavelet images in combination with discriminant analysis, we obtained classification accuracies of 96.67 and 87.78% for the training and testing set (90 images each), respectively. We conclude that the best classifier of carcinomas in histological sections of breast tissue are the texture features from the second-level Haar transform wavelet images used in a discriminant function.