Non-Rigid Surface Registration (2008)

We proposed a new method for non-rigid surface registration between a surface model and a surface of an internal organ in a given 3D medical image. The surface is represented with a set of feature points, of which locations are represented by a graphical model. For constructing the representation, a set of corresponding points is distributed on each of training surfaces based on an entropybased particle system. From these corresponding points, we estimate probability densities of the location of each feature point, the conditional probability distribution of the local image pattern around each feature point, and the probability distributions of relative positions between two neighboring feature points. When a new image is given, these densities are used for estimating the location of each feature point by means of a non-parametric belief propagation. The proposed method can estimate not only the locations of the feature points but also their conditional marginal distributions in a given image. Some experimental results obtained from real X-CT images are presented to show its performance.

Unbiased Logan Graphical Analysis (2008)

The Logan Graphical Analysis (LGA) is used for imaging a distribution volume VT. As known, LGA underestimates VT and its unbiased estimator is expected. We applied Renormalization Method (RM) for LGA and investigated its applicability.

Automatic Construction of Statistical Model of Calcified Aorta Areas (2007)

We reported on automatic construction of a statistic model of calcified aorta areas. For the model construction, we developed a method for accurately detecting aorta areas and calcified ones from noncontrasted X-ray CT images. The proposed method firstly extracts the medial line of aorta in a given image by means of the active contour. Secondly the method extracts the wall of the aorta accurately by means of the Tensor Voting. Our method thirdly extracts calcified areas in the extracted aorta, and finally integrates the extracted areas in an aorta model.

Successive Localization of Mobile Sensor Nodes Using Non-Parametric Belief Propagation (2006)

We proposed a method for positioning networked sensor nodes that include mobile ones. When nodes can measure the relative distances to their neighbor ones, we can estimate the absolute position of each node. Once distributed static nodes are positioned, we can position a mobile node for each time that moves through the distributed ones and measures the relative distances to the distributed ones. By using these distances, we can improve the estimated positions of the distributed nodes. For this estimation, we employ the non-parametric belief propagation. Simulation results showed that the proposed method improved the estimated positions.

Stability Evaluation of a Classifier for Detecting Abdominal Tumors in FDG-PET/CT Images (2006)

The purpose of our study was to develop a completely automated method for detecting abdominal tumors in given FDG-PET/CT images. The purpose of this article is to report on a classier that distinguishes malignant tumors from false positives. For constructing the classier, we employed the kernel Fisher discriminant analysis (KFDA), and experimentally evaluated the relationship between the dimensionality of training data, the leave-one-out (LOOCV) error of the classier, and the stability of the classier. The results showed that, as the dimensionality increased, the LOOCV error decreased but the classier became unstable. Our method rstly binarizes a given PET image to extract candidates of malignant tumors. From a set of the images of the extracted candidates, we constructed the classier. For computing the classication function, we normalize the size of the extracted images of the candidates and incorporate them into training data. When we change the size of the normalized images, the dimensionality of the training data changes and we obtain different classication function. For each classi- cation function obtained at different dimensionality, we approximately evaluated its uniform stability and computed the upper bound of the difference between the LOOCV error of the classier and the generalization error. In this article, we show that it is needed to evaluate the stability of the classier to determine the size of the normalized images.

Improvement of Vehicle Positioning Using Car-to-Car Communications in Consideration of Communication Delay (2005)

We proposed a car positioning method that utilizes car-to-car communications. When each car can measure the relative distances between neighboring cars and can send the measurements to those cars, we can improve the accuracy of the estimated car locations by means of the Kalman filtering. Here, the problem is the delay in the communication. We handled this problem and proposed a new method for the improvement.