We developed upright CT for whole-body scan. It has the same physical characteristics as a conventional CT, requires only 2/3 of the space of a conventional CT, and can be performed with a workflow similar to that of radiography. We have revealed that the brain and pelvis of healthy volunteer, which were previously thought to remain unchanged in position, descend in the upright position, gradually revealing the structure of the human body under gravity. In particular, the veins show different capacitance changes between the upright and supine positions depending on the area. It is also possible to evaluate the muscle volume of each area, such as pelvic floor laxity, the trunk and thighs, and to suggest ways to exercise according to individual muscle atrophy. As for diseases, hernias and organ prolapse can be evaluated quantitatively, and early detection of musculoskeletal diseases such as knee osteoarthritis is also possible. Furthermore, dynamic functions such as swallowing, respiration, urination, and gait can be evaluated and related pathologies can be detected at an early stage. Through upright CT, we hope to contribute to the needs of an era of health and longevity.

We developed an original magnetic resonance imaging (MRI) system that can obtain images in multiposture (Gravity MRI). We described research results of the Gravity MRI of the central nervous system, internal organs, and musculoskeletal system. Gravity MRI makes it possible to noninvasively obtain new functional and morphological information.

What is the ideal brain-dedicated PET system? Our answer to this question is a hemispherical detector arrangement. This geometry, which may fit the head the best, improves the spatial resolution as well as the sensitivity without increasing the number of detectors. After seven years of research and development, we developed the worldʼs first hemispherical brain PET “VRAIN”. A sitting style is also a major feature of VRAIN, which may improve patient comfort. A reduced footprint could help hospitals save space.

Musculoskeletal models that are faithful to medical images are important for clarifying the causes of movement disorders. Since the kinematic properties of the musculoskeletal system change depending on the body position, it is desirable to create a musculoskeletal model of standing movements based on medical images scanned in the standing position. However, medical images scanned in the upright position are less common than those in the supine position, and data are limited. Therefore, the final goal of our project is to quantify musculoskeletal deformation in the upright position and predict it from the images scanned in the supine position in order to effectively use images scanned in the supine position. In this study, as a first step, we automatically segmented supine and upright CT images scanned for the same subject, and determined the segmentation accuracy with respect to the body position.

Quantitative susceptibility mapping (QSM) is promising for the diagnosis of diseases in which the magnetic sus ceptibility of tissues changes; QSM is mainly performed on 3T MRI. In this study, with the aim of practical application of QSM in 1.5T MRI, the relationship between QSM noise and imaging parameters (TR/TE) was compared between numerical calculations and actual measurements. 1.5T QSM phantom experimental results showed good agreement with the calculations. The results of the phantom experiment of 1.5T QSM agreed well with the calculations, and the same results were obtained in the measurement of the human brain in healthy subjects. The QSM values of the human brain were consistent with those reported in previous 3T MRI studies. These results suggest the feasibility of 1.5T QSM.

Parallel imaging in magnetic resonance imaging produces non-uniform noise when the acceleration factor is high. To reduce non-uniform noise, we developed a method that combines iterative reconstruction with soft-thresholding and CNN-based denoising. Non-uniform noise is reduced by iterative reconstruction, and further reduced by combining iterative reconstruction with CNN-based denoising. We applied this method to parallel imaging of the heads of six volunteers and confirmed that our method was able to reduce non-uniform noise in reconstructed images with an acceleration factor of 3, where the g-factor is high. Compared with a conventional reconstruction with CNN-based denoising for an acceleration factor of 3, our method improved the structural similarity index measure values relative to the reference images for T2-weighed, T1-weighted, and FLAIR images from 0.922/0.929/0.864 to 0.932/0.938/0.886, respectively (P < 0.05). Therefore, our method contributes to improving image quality with a shorter scan time.