The utility of lacrimal passage contrasting by a digital subtraction system (DS system) was assessed in comparison with a computed radiography system (CR system) by means of simulating the exposure dose of a patient’s crystalline lens and measuring the image contrast of both systems. The exposure dose of the patient’s crystalline lens in the DS system was an average of 45.8 mGy, which was 41.6 times higher than that of the CR system. Therefore, care must be taken to reduce the exposure dose of the crystalline lens because it is necessary to reduce the probability of radiation injuries such as cataracts. The average of the image contrast of the DS system at the lacrimal passage to other parts of the head radiograph was lower than that of the CR system, but the standard deviation of the DS system was 0.16, a value that was almost constant because the shadow of obstacles such as the facial bone was almost completely removed, and image contrast was improved. The area under the curve (AUC) of the DS and CR systems as determined by means of receiver operating characteristic (ROC) analysis by ten radiological technologists were 0.869±0.066 and 0.746±0.125, respectively, and statistical significance was shown for both systems, although the detectability of the DS system was superior to that of the CR system (p<0.05). Therefore, we concluded that the DS system was a useful radiographic technique for lacrimal passage contrasting, and its use in patients is predicted to increase in the future.
The purpose of this study was to conduct fast-acquired muscle functional magnetic resonance imaging (fast-mfMRI). Fast-mfMRI is a method of fusing fast MR images in order to visualize muscle activity. Exercise selectively increases the signal intensities (SI) of active muscles in T2-weighted magnetic resonance (MR) images. A fast-mfMRI image is a fusion of two types of images: an anatomic image acquired by the TrueFISP method and a functional image acquired by the SE-EPI (spin-echo echo-planar-imaging) method. MR images of four healthy males were recorded at rest before and after plantar flexion. The Gain of the MR signal remained constant from before the flexion exercise (at rest) to after the exercise. The data on the area of muscle activity could be extracted by adapting a threshold value obtained by a functional image at rest to the functional image after the exercise. By uniting the data on the area of muscle activity with the anatomic images after the exercise, we constructed a fused image rich in anatomical information and effective in visualizing muscle activity. These fast-mfMRI images can be acquired in 14 seconds. Our results suggest that fast-mfMRI has the potential to measure muscle activity in the trunk, where conventional mfMRI has been ineffective.
We investigated and evaluated the detection of simulated lesions in various interstitial lung diseases using the dual-energy subtraction radiography method and flat-panel detector (FPD) images. We obtained a FPD system (GE Revolution XR/d), and employed dual-energy 60 kV and 130 kV exposure techniques. Three types of lung lesions, namely, micro-nodule, ground-glass, and honeycomb patterns were simulated with interstitial lung disease on a chest phantom. Chest images with and without simulated lesions were exposed and compared with standard images and subtraction images. We carried out evaluations with and without subtraction images and performed the analysis by using receiver operating characteristic (ROC) analysis of detection. Results showed that the detection of interstitial lung diseases was significantly improved by the use of subtraction images. The area under the ROC curve (AUC) values of micro-nodule images obtained with and without subtraction images were 0.768 and 0.963, ground-glass images 0.670 and 0.917, and honeycomb images 0.768 and 0.996, respectively. A significant difference of p<0.05 was accepted. The use of dual-energy subtraction radiography with a FPD improved diagnostic accuracy in detecting cases of multiple interstitial lung diseases and was considered useful.
In recent years, the advancements in MR technology combined with the development of the multi-channel coil have resulted in substantially shortened inspection times. In addition, rapid improvement in functional performance in the workstation has produced a more simplified imaging-making process. Consequently, graphical images of intra-cranial lesions can be easily created. For example, the use of three-dimensional spoiled gradient echo (3D-SPGR) volume rendering (VR) after injection of a contrast medium is applied clinically as a preoperative reference image. Recently, improvements in 3D-SPGR VR high-resolution have enabled accurate surface images of the brain to be obtained. We used stereo-imaging created by weighted maximum intensity projection (Weighted MIP) to determine the skin incision line. Furthermore, the stereo imaging technique utilizing 3D-SPGR VR was actually used in cases presented here. The techniques we report here seemed to be very useful in the pre-operative simulation of neurosurgical craniotomy.
Determination of the region of interest (ROI) for dynamic renal function has been highly discordant among operators because of the dependence on factors such as the rate of injection of radioactive medicines, constitution, and renal function. To simplify this problem, we developed a computer algorithm that provides automatic analysis for both localization of the kidney and automatic determination of ROIs using computed tomographic (CT) images. The bilateral kidneys were extracted from enhanced CT images, and the CT pixel size was adjusted to dynamic renal function images. The template-matching technique was used for these images, and the kidney renal location was analyzed on additional functional images constructed by dynamic renal function images. As a result, we were able to obtain time-activity curves of both renal function and quantified glomerular filtration rate (GFR). In conclusion, the computer algorithm we developed was considered to provide reliable results, apart from the variability among operators, because of its good reproducibility.
Classification of the liver region of the Couinaud segment provides significant information for a computer-aided diagnostic system to localize the position of lesions in the liver region. Hepatic vessels provide essential information to classify the liver region of the Couinaud segment. However, automated segmentation and classification of hepatic vessels are difficult in non-contrast CT images owing to the low contrast between hepatic vessels and liver tissue. In this paper, we propose an automated extraction schema for extracting the middle hepatic vein (MHV), and we employ this schema to classify the liver region into right and left lobes. We applied our method to 22 non-contrast X-ray CT images. All of the cases were normal liver cases. The results for the MHV extraction were evaluated using three parameters for the volume ratio to the correct region of liver. The results show that hepatic vessels extracted using the proposed method were found to be satisfactory in 41% (9/22) of cases.