Masahide Nishiura

A semiautomated technique for evaluation of uterine peristalsis.

To evaluate the capability of a newly developed semiautomatic analysis technique for evaluation of uterine peristalsis in comparison with visual assessment.

Robust lip contour extraction using separability of multi-dimensional distributions

We present a lip contour extraction method using separability of color intensity distributions. Usually it is difficult to robustly extract the outer lip contour mainly because of the following two problems. First, the outer lip contour is often blurred. Secondly, the contrast between the skin and the lip region is often reduced by transformation from the color intensity to the gray scale intensity. To overcome these two problems we propose an edge detection method in which edge strength is defined as separability of two color intensity distributions. We apply the proposed method to lip contour extraction using an active contour model. We present several experimental results demonstrating the effectiveness of the proposed method.

Active contour extraction method using partial shape constraint contour model

In this paper, we propose a contour extraction method that uses the partial shape constraint contour model and is based on knowledge about the structure of the heart. This method detects the positions of annuli at the bases of the valves in a tomographic ultrasound image of the left ventricle from image patterns and creates a contour model under partial constraints based on these positions. Because a model that constrains the partial position of a contour is used instead of the conventional statistical model of the entire contour, contours can be accurately extracted even for statistically few shapes and without having to learn the contour shapes. The proposed method uses clinical images to show that it is possible to reduce incorrect contour extractions compared to when the partial shape constraint contour model is not used.

Fully automatic feature localization for medical images using a global vector concentration approach

In this paper, we propose a novel feature localization method based on a global vector concentration approach. Our approach does not rely on the detection of local salient features around feature points. Instead, we exploit global structural information of the object extracted by calculating the concentration of directional vectors from sampling points. Those vectors are combined with local pattern descriptors of a query image and selected from preliminarily trained extended templates by nearest neighbor search. Due to the insensitivity of local changes, our method can handle partially occluded and noisy objects. We apply the proposed method to fully automatic feature localization of the left ventricular in echocardiograms. The results show the effectiveness of our method in comparison with a conventional edge-based method in terms of accuracy and robustness.

Automatic heart wall contour extraction from MR images using active contour models: Initial contour setting based on principal component analysis

For the problem of initial value setting, which will become a problem in contour extraction using active contour models, we propose in this paper a scheme for effectively inputting the initial values by facilitating user input with statistical models of the contour shapes of the heart wall. We also describe a scheme for extracting and tracking the left ventricular contour of the heart from the MR images by the edge detection method adapted to low-contrast images. Moreover, we will show the validity of the proposed scheme by comparison with the manually traced results of physicians.