Normalization correction is necessary to obtain high-quality reconstructed images in positron emission tomography (PET). There are two basic types of normalization methods: the direct method and component-based methods. The former method suffers from the problem that a huge count number in the blank scan data is required. Therefore, the latter methods have been proposed to obtain high statistical accuracy normalization coefficients with a small count number in the blank scan data. In iterative image reconstruction methods, on the other hand, the quality of the obtained reconstructed images depends on the system modeling accuracy. Therefore, the normalization weighing approach, in which normalization coefficients are directly applied to the system matrix instead of a sinogram, has been proposed. In this paper, we propose a new component-based normalization method to correct system model accuracy. In the proposed method, two components are defined and are calculated iteratively in such a way as to minimize errors of system modeling. To compare the proposed method and the direct method, we applied both methods to our small OpenPET prototype system. We achieved acceptable statistical accuracy of normalization coefficients while reducing the count number of the blank scan data to one-fortieth that required in the direct method.

View full abstract

Practical CT reconstruction algorithms incorporate assumptions concerning the X-ray source trajectory, and the detectors rotate in a circular trajectory. However, in order to reduce the scanning time or to reconstruct objects that move during projection acquisition, a free-form X-ray source trajectory is needed. In this paper, we present the derivation of a general reconstruction formula for a free-form trajectory from fan-beam projection data. Furthermore, we have rewritten the reconstruction formula using both equal-angle rays and equally spaced collinear detectors for practical implementation and application. As in any reconstruction processing, it is necessary to address the redundant projection data problem. We propose a plus-minus weighting scheme that avoids the need to calculate the number of intersection points between each X-ray line and trajectory. A numerical study was conducted using a standard Shepp-Logan phantom to verify the proposed algorithm.

View full abstract

Rupture of the Achilles tendon is a common injury during sports activities. Since this injury can cause many problems in daily life, it would be of great value to develop a method for predicting rupture of the Achilles tendon. The aim of our research is to detect subtle signs of impending rupture noninvasively based on multiple MR images. In the present study, we investigated the angle dependence of the T₂ relaxation time of collagen fibers (MAE: Magic Angle Effect). The proposed method is based on the acquisition of multiple MR images with various angles between the static magnetic field and the fiber orientation. By fitting a theoretical curve to the signal intensity profile, it is possible to calculate parameters that can be used for the quantitative evaluation of impending Achilles tendon rupture. The results showed that the proposed method is useful for detecting areas of disordered collagen fiber orientation.

View full abstract

The number of patients with cardiomegaly has been increasing in recent years. Cardiomegaly is a condition in which the heart is enlarged, resulting in an increase in the cardiothoracic ratio. The heart is subjected to an increased workload, which ultimately leads to heart failure, ischemic heart disease, dilated cardiomyopathy, etc. The early detection of cardiomegaly is essential for the treatment of such pathological conditions. In methods based on plain X-ray images, the condition is identified by directly measuring the cardiothoracic ratio on the image with a ruler. However, this has been found to be an inaccurate method for identifying cardiomegaly due to the large differences caused by variation in the X-ray exposure conditions. In this report, we propose a method in which cardiomegaly is identified based on the volumetric ratio of the heart and lung field volumes (called the cardiothoracic volumetric ratio) extracted from computed tomography (CT) images. Regional extraction of each organ is performed by setting a seed point for extraction in both the heart and the lungs and then performing labeling and morphological processing. To evaluate the usefulness of this method, CT images of eight test subjects, three of whom were diagnosed as having cardiomegaly, were used. The results showed a clear difference between the three subjects with cardiomegaly and the other five subjects, with average cardiothoracic volumetric ratios of 32% in the subjects with cardiomegaly and 15% in the other subjects. These findings indicate that the proposed method should be useful for the identification of cardiomegaly.

View full abstract

[in Japanese]

View full abstract