Although it has been 40 years since MRI equipped with gradient coil was developed by Dr. Lauterbur in 1973, MRI technology is still under development for both hardware and imaging methods, and has been commercially available. This article introduces two current technologies, parallel transmission available on Siemens MRI and multi-band EPI under development. Parallel transmission can reduce transmission field inhomogeneity, which is caused by penetration depth of high frequency Radio frequency, typically on 3T MRI. Moreover, parallel transmission enables the local excitation along with read-out and phase-encoding direction, in addition to the excitation along with slice direction in the conventional method. Multi-band EPI, one of outcomes of Human Connectome Project in the USA, can be widely applied to fMRI (functional MRI), Diffusion Tensor Imaging (DWI) and Diffusion Spectrum Imaging (DSI), offering high-temporal and spatial resolution, reduction of data acquisition time, and improvement of statistical power.

View full abstract

The MR signal consists of magnitude and phase information. In late years various clinical applications that utilized phase information are reported. In this article, we describe about representative imaging processing methods (phase contrast MRA, diffusion weighted imaging, functional MRI (fMRI), susceptibility weighted imaging) which utilized phase information.

View full abstract

In recent years, importance of holistic solution integrating information from US, CT and MRI are increasing for commercial medical image diagnostics and treatment techniques. In this paper, we overview the latest image processing applications for early detection and treatment of liver cancer across modalities from screening, diagnosis, treatment planning to follow-up inspection. Expectation for the future development are described.

View full abstract

This paper shows research and development of advanced clinical applications of Toshiba groups. Toshiba groups have been researching and developing applications which include not only image processing technologies but also scanner system technologies based on clinical needs, business needs, and technology seeds. In this paper, planning assist technologies for cardiac magnetic resonance imaging examinations are presented as one example of the applications.

View full abstract

We have developed "SYNAPSE Case Match", content-based image retrieval system for supporting Lung Cancer image diagnosis, in collaboration with Shizuoka Cancer Center, and started commercialization in October 2012. When performing the interpretation of lung cancer in the CT inspection image, this system can search the past similar cases from the case database, and display them in the order of similarity. The imaging diagnosis work requires a comprehensive judgment based on medical knowledge and experience. This system is expected to effectively support the interpretation of physician.

View full abstract

Pleural effusion has a number of causes, such as cardiac failure, liver cirrhosis, pneumonia and cancer. When increased pleural fluid, pneumothorax or other intrapleural space-occupying lesions press on the bronchial tubes, they develop atelectasis in their peripheral areas. Both pleural fluid and atelectasis potentially cause dyspnea, with atelectasis causing a concurrence of pneumonia or other condition, necessitating early treatment. In conventional diagnosis using plain X-ray images, it is extremely difficult to distinguish between adjoining pleural fluid and compressive atelectasis regions. In this study, a method is proposed by which chest contrasting computed tomography images are used to differentiate regions of pleural fluid and compressive atelectasis from other regions and measure individually the volume of each region by semiautomatic processing. Quantification enables numerical assessment of the progress of the disease condition and of treatment. Lung air space region extraction, lung contour detection from rib information, and heart region removal are performed, the resulting images of pleural fluid and compressive atelectasis are subjected to close examination, and volumes of specific areas are measured. The results of verification using patient data suggested the effectiveness of quantitative measurement of specific areas using this method.

View full abstract

When defining a complexity as the number of dominant movement patterns of the lung internal structures movement, a variety of different situations are found in COPD (chronic obstructive pulmonary disease) patients. Furthermore, there is no clear correspondence among the complexity and the disease severity. On the other hand, the LAA% (ratio of low attenuation area to total lung area) is reported to correlate with respiratory function parameters, such as FEV1/FVC (ratio of forced expiratory volume in 1 second to forced vital capacity) to some degree. This study attempts to explain this diversity by combining dynamic parameters derived from the internal lung structures movement with the LAA%. Time sequential MR sagittal images with balanced FFE sequences are taken at the mid portion of the right lung during deep breathing for 5 healthy volunteers and 24 patients with COPD (4 in Grade I; 10 in Grade II, 8 in Grade III and 2 in Grade IV).CT images were also taken during breath holding. The movement of the lung is extracted from MR images by tracking sample points on arterial and venous blood vessels up to subsegmental branches. Main motion patterns are obtained for each lobe as well the whole lung. Dynamic parameters such as variance of displacement and swept area are obtained for each lobe from the sample points movement. LAA% is obtained from CT images. The correlations between FEV1/FVC and parameters derived from the lung motion as well as LAA% are performed using multi-regression analysis. The conclusions are that LAA% contribution to FEV1/FVC is practically equal to the lung movement dispersions and that a high correlation is obtained from the combination of LAA% and dynamic parameters.

View full abstract

An ultrasonic contrast echo method using two beams with different frequencies is described. Summation and difference components in the echo signals received from the microbubbles are generated. Thus, the blood flow is estimated by the Dopplar frequency of those components. This method has advantages of SNR improvements and absolute value measurement.

View full abstract