Prediction of endovascular leaks after thoracic endovascular aneurysm repair though machine learning applied to pre-procedural computed tomography angiographs.

To predict endoleaks after thoracic endovascular aneurysm repair (TEVAR) we submitted patient characteristics and vessel features observed on pre- operative computed tomography angiography (CTA) to machine-learning. We evaluated 1-year follow-up CT scans (arterial and delayed phases) in patients who underwent TEVAR for the presence or absence of an endoleak. We evaluated the effect of machine learning of the patient age, sex, weight, and height, plus 22 vascular features on the ability to predict post-TEVAR endoleaks. The extreme Gradient Boosting (XGBoost) for ML system was trained on 14 patients with- and 131 without endoleaks. We calculated their importance by applying XGBoost to machine learning and compared our findings between with those of conventional vessel measurement-based methods such as the 22 vascular features by using the Pearson correlation coefficients. Pearson correlation coefficient and 95% confidence interval (CI) were r = 0.86 and 0.75 to 0.92 for the machine learning, r = - 0.44 and - 0.56 to - 0.29 for the vascular angle, and r = - 0.19 and - 0.34 to - 0.02 for the diameter between the subclavian artery and the aneurysm (Fig. 3a-c, all: p < 0.05). With machine-learning, the univariate analysis was significant higher compared with the vascular angle and in the diameter between the subclavian artery and the aneurysm such as the conventional methods (p < 0.05). To predict the risk for post-TEVAR endoleaks, machine learning was superior to the conventional vessel measurement method when factors such as patient characteristics, and vascular features (vessel length, diameter, and angle) were evaluated on pre-TEVAR thoracic CTA images.

Comparison of sensitivity characteristics between GAFCHROMIC LD-V1 and GAFCHROMIC XR-QA2 in mammographic dosimetry.

In recent years, radiochromic films have begun to be used for dosimetry in mammography; however, the most sensitive GAFCHROMIC XR-QA2 (XR-QA2) film is no longer available owing to its discontinuation. In this study, we evaluated the sensitivity characteristics of GAFCHROMIC LD-V1 (LD-V1) as an alternative to XR-QA2 in the field of mammography, at a low dose and low energy. Our results show that the average ratio of the concentration change of LD-V1 divided by the concentration change of XR-QA2 at each absorbed dose was 53.7%, indicating the sensitivity of LD-V1 to be approximately half of XR-QA2. In addition, the linearity of the concentration change is sufficient even within a dose range of 0.59-14.52 mGy, which is lower than the manufacturer's recommended dose range. Therefore, the LD-V1 is capable of accurate dose assessment even with a low dose and the low level of energy used in mammography.

Usefulness of electrocardiogram mA modulation during the electrocardiogram-gated CT scan in paediatrics with high heart rate for different helical pitch: a phantom-based assessment study.

We investigated the effect of electrocardiographic (ECG) mA-modulation of ECG-gated scans of computed tomography (CTA) on radiation dose and image noise at high heart rates (HR) above 100 bpm between helical pitches (HP) 0.16 and 0.24. ECG mA-modulation range during ECG-gated CTA is 50-100 mA, the phase setting is 40-60% and the scan range is 90 mm for clinical data during HR for 90, 120 and 150 bpm. Radiation dose and image noise in Housfield units are measured for CT equipment during HR for 90, 120 and 150 bpm between HP 0.16 and 0.24. ECG mA-modulation, dose reduction ratio for HR 90, 120 and 150 bpm are 19.1, 13.4 and 8.7% at HP 0.16 and 17.1, 13.3 and 7.7% at HP 0.24, respectively. No significant differences were observed in image noise between both HP. Dose reductions of 8-24% are achieved with ECG mA-modulation during ECG-gated CCTA scan, which is beneficial even in high HR more than 100 bpm.

RADIATION DOSE REDUCTION AT LOW TUBE VOLTAGE WITH CORONARY ARTERY BYPASS GRAFT COMPUTED TOMOGRAPHY ANGIOGRAPHY BASED ON THE CONTRAST NOISE RATIO INDEX.

To compare the radiation dose and diagnostic ability of the 100-kVp protocol, based on the contrast noise ratio (CNR) index, during coronary artery bypass graft (CABG) vessels with those of the 120-kVp protocol. For the 120-kVp scans (150 patients), the targeted image level was set at 25 Hounsfield units (HU) (CNR120 = iodine contrast/25 HU). For the 100-kVp scans (150 patients), the targeted noise level was set at 30 HU to obtain the same CNR as in the 120-kVp scans (i.e. using 1.2-fold higher iodine contrast, CNR100 = 1.2 × iodine contrast/(1.2 × 25 HU) = CNR120). We compared the CNRs, radiation doses, detection of CABG vessels and visualisation scores of the scans acquired at 120 and 100 kVp, respectively. At the same CNR, the 100-kVp protocol may help reduce the radiation dose by ⁓30% compared with the 120-kVp protocol, without degradation of diagnostic ability during CABG.

Effect of patient characteristics on vessel enhancement on arterio-venous fistula CT angiography in a retrospective cohort study.

To evaluate the effects of various patient characteristics on vessel enhancement on arterio-venous fistula (AVF) computed tomography (CT) angiography (AVF-CT angiography). A total of 127 patients with suspected or confirmed shunt stenosis and internal AVF complications were considered for inclusion in a retrospective cohort study. The tube voltage was 120 kVp, and the tube current was changed from 300 to 770 mA to maintain the image quality (noise index: 14) using automatic tube current modulation. To evaluate the effects of age, sex, body size, and scan delay on the CT number of the brachial artery or vein, we used correlation coefficients and multivariate regression analyses. There was a significant positive correlation between the CT number of the brachial artery or vein and age (R = 0.21 or 0.23, P < .01). The correlations were inverse with the height (r = -0.45 or -0.42), total body weight (r = -0.52 or -0.50), body mass index (r = -0.21 or -0.23), body surface area (body surface area [BSA]; r = -0.56 or -0.54), and lean body weight (r = -0.55 or -0.53) in linear regression analysis (P < .01 for all). There was a significant correlation between the CT number of the brachial artery or vein and scan delay (R = 0.19 or 01.9, P < .01). Only the BSA had significant effects on the CT number in multivariate regression analysis (P < .01). The BSA was significantly correlated with the CT number of the brachial artery or vein on AVF-CT angiography.

Development of a three-dimensional dose evaluation method for computed tomography.

During a single scan using computed tomography, an X-ray tube orbits along a 360°-circular path around the patient. A scan obtained using the half-cylindrical type phantoms with a radiochromic film sandwiched in between reveals a pixel value map illustrating the two-dimensional (2D) dose distribution. A three-dimensional (3D) dose distribution can be obtained with a 360° rotation of the 2D dose map. This study evaluates the concept and methodology of creating a 3D dose map to develop a phantom with a radiochromic film for obtaining the 3D dose distribution. The coronal and axial plane dose distributions were also evaluated. A single scan computed tomography image obtained using a half-cylindrical type of acrylic phantom with a sandwiched radiochromic film was studied. The diameters of the phantoms were 10 and 16 cm, and their lengths were 30 cm. A 2D image of the XR-QA2 film was obtained using an image scanner and image processing software. A red channel image was used to obtain the 3D dose distribution using a computing platform. A pseudo color was applied to the red channel image from which cross-sectional color images were obtained. Half of the cross-sectional pixel data were rotated by 360° to generate the data for each axial plane. The image created was saved, and a 3D pixel value map was constructed. The dose measurement procedure for the 3D dose distribution was developed using half-cylindrical acrylic phantoms with a radiochromic film.

Usefulness of the patient-specific contrast enhancement optimizer simulation software during the whole-body computed tomography angiography.

To evaluate whether the patient-specific contrast enhancement optimizer simulation software (p-COP) is useful for predicting contrast enhancement during whole-body computed tomography angiography (WBCTA). We randomly divided the patients into two groups using a random number table. We used the contrast material (CM) injection protocol selected by p-COP in group A (n = 52). The p-COP used an algorithm including data on the individual patient's cardiac output. Group B (n = 50) was assigned to the conventional CM injection protocol based on body weight. We compared the CT number in the abdominal aorta at the celiac artery level between the two groups and classified them as acceptable (> 280 HU) and unacceptable (< 279 HU) based on the optimal CT number for the WBCTA scans. To evaluate the difference in both injection protocols, we compared the visual inspection of the images of the artery of Adamkiewicz in both protocols. The CM dosage and injection rate in group A were significantly lower than those in group B (480.8 vs. 501.1 mg I/kg and 3.1 vs. 3.3 ml/s, p < 0.05). The CT number of the abdominal aorta at the celiac level was 382.4 ± 62.3 HU in group A and 363.8 ± 71.3 HU in group B (p = 0.23). CM dosage and injection rate were positively correlated to cardiac output for group A (r = 0.80, p < 0.05) and group B (r = 0.16, p < 0.05). The number of patients with an acceptable CT number was higher in group A [46/6 (86.7%)] than in group B [43/7 (71.4%)], but not significant (p = 0.71). The visualization rate for the Adamkiewicz artery was not significantly different between groups A and B (p = 0.89). The p-COP was useful for predicting contrast enhancement during WBCTA with a lower CM dosage and a lower contrast injection rate than that based on the body weight protocol. In patients with lower cardiac output a reduction in contrast injection rate and CM dosage did not lead to a reduced imaging quality, thus particularly in this group CM dosage can be reduced by p-COP.

THREE-DIMENSIONAL DOSIMETRY OF MAMMOGRAPHY USING A BLOCK CUBE BREAST PHANTOM AND RADIOCHROMIC FILM.

This study developed a phantom with a shape similar to that of the breast and use GAFCHROMIC films that can be placed inside the phantom to measure the detailed breast dose distribution in mammography. GAFCHROMIC EBT3 was placed on the block cube breast phantom and irradiated with a mammography device to measure the absorbed dose distribution inside the phantom in the horizontal and depth directions. The dose distribution in the horizontal plane was the highest in the centre on the chest wall side, and it decreased in a fan shape. Along the depth of the phantom, the doses absorbed across the entire cross-section were 16.15 mGy at the surface and 7.51, 3.25 and 1.68 mGy at depths of 10, 20 and 30 mm, respectively. Compared with the mean glandular dose, the proposed method can measure breast dose distributions in greater detail and is applicable to various breast shapes.

USE OF ULTRAVIOLET RAY PRE-IRRADIATION TO IMPROVE THE ACCURACY OF LOW-DOSE MEASUREMENTS OF THE CT USING A GAFCHROMIC RTQA2 FILM.

The purpose of this study is to develop a method for use at extremely low-dose ranges and to decrease the uncertainty outside the recommended range of Gafchromic RTQA2 (RTQA2). By this method, the CT dose including the scattered radiation region can be grasped. The base density was increased by ultraviolet (UV)-ray preirradiation. RTQA2 was irradiated with UV-A rays for 26 and 40 h. Subsequently, RTQA2 was exposed to 2, 4, 6, 8, 10, 25, 50, 75, 100, 150, 200 and 250 mGy X-rays using a segmentation method. Calibration curves with and without UV-A irradiation were compared. The calibration curve with 40-h UV-A ray irradiation was the most linear, and a steeper slope area was not observed. The uncertainty in the calibration curve was reduced (p < 0.05). UV-A ray irradiation is an effective method for treating RTQA2; the accuracy in the extremely low-dose range of RTQA2 was improved.

THE CONCEPT OF X-RAY CT DOSE EVALUATION METHOD USING RADIOCHROMIC FILM AND FILM-FOLDING PHANTOM.

In this paper, we propose a novel radiochromic film (RCF)-based computed tomography (CT) dosimetry method, which is different from the method based on CT dose index. RCF dosimetry using Gafchromic QA2 films was performed using two lengths of film-folding phantoms. The phantom was exposed to X-ray CT through a single scan, while the RCF was sandwiched between the phantoms. We analysed the dose profile curve in two directions to investigate the dose distribution. We observed a difference in the dose distribution as the phantom size changed. Our results contradict with the results of previous studies such as Monte Carlo simulation or direct measurement. The ability to visually evaluate 2D dose distributions is an advantage of RCF dosimetry over other methods. This research investigated the ability of 2D X-ray CT dose evaluation using RCF and film-folding phantom.

Intensities of Incident and Transmitted Ultraviolet-A Rays through Gafchromic Films.

Gafchromic films have been applied to X-ray dosimetry in diagnostic radiology. To correct nonuniformity errors in Gafchromic films, X-rays in the double-exposure technique can be replaced with ultraviolet (UV)-A rays. Intensities of the incident and transmitted UV-A rays were measured. However, it is unclear whether the chemical color change of Gafchromic films affects the UV-A transmission intensity. Gafchromic EBT3 films were suitable to be used in this study because non-UV protection layers are present on both sides of the film. The film is placed between UV-A ray light-emitting diodes and a probe of a UV meter. Gafchromic EBT3 films were irradiated by UV-A rays for up to 60 min. Data for analysis were obtained in the subsequent 60 min. Images from before and after UV-A irradiation were subtracted. When using 375 nm UV-A, the mean ± standard deviation (SD) of the pixel values in the subtracted image was remarkably high (11,194.15 ± 586.63). However, the UV-A transmissivity remained constant throughout the 60 min irradiation period. The mean ± SD UV-A transmission intensity was 184.48 ± 0.50 µm/cm2. Our findings demonstrate that color density changes in Gafchromic EBT3 films do not affect their UV-A transmission. Therefore, Gafchromic films were irradiated by UV-A rays as a preexposure.

Correction of nonuniformity error of Gafchromic EBT2 and EBT3.

This study investigates an X-ray dose measurement method for computed tomography using Gafchromic films. Nonuniformity of the active layer is a major problem in Gafchromic films. In radiotherapy, nonuniformity error is reduced by applying the double-exposure technique, but this is impractical in diagnostic radiology because of the heel effect. Therefore, we propose replacing the X-rays in the double-exposure technique with ultraviolet (UV)-A irradiation of Gafchromic EBT2 and EBT3. To improve the reproducibility of the scan position, Gafchromic EBT2 and EBT3 films were attached to a 3-mm-thick acrylic plate. The samples were then irradiated with a 10 W UV-A fluorescent lamp placed at a distance of 72cm for 30, 60, and 90 minutes. The profile curves were evaluated along the long and short axes of the film center, and the standard deviations of the pixel values were calculated over large areas of the films. Paired t-test was performed. UV-A irradiation exerted a significant effect on Gafchromic EBT2 (paired t-test; p = 0.0275) but not on EBT3 (paired t-test; p = 0.2785). Similarly, the homogeneity was improved in Gafchromic EBT2 but not in EBT3. Therefore, the double-exposure technique under UV-A irradiation is suitable only for EBT2 films.

Simplified method for creating a density-absorbed dose calibration curve for the low dose range from Gafchromic EBT3 film.

Radiochromic film dosimeters have a disadvantage in comparison with an ionization chamber in that the dosimetry process is time-consuming for creating a density-absorbed dose calibration curve. The purpose of this study was the development of a simplified method of creating a density-absorbed dose calibration curve from radiochromic film within a short time. This simplified method was performed using Gafchromic EBT3 film with a low energy dependence and step-shaped Al filter. The simplified method was compared with the standard method. The density-absorbed dose calibration curves created using the simplified and standard methods exhibited approximately similar straight lines, and the gradients of the density-absorbed dose calibration curves were -32.336 and -33.746, respectively. The simplified method can obtain calibration curves within a much shorter time compared to the standard method. It is considered that the simplified method for EBT3 film offers a more time-efficient means of determining the density-absorbed dose calibration curve within a low absorbed dose range such as the diagnostic range.

Ultraviolet exposure of Gafchromic XR-RV3 and XR-SP2 films.

Gafchromic film has been used for X-ray dose measurement in diagnostic examinations. Their use has been initiated for three-dimensional X-ray dose measurement by using the high-resolution characteristics of Gafchromic films in computed tomography. However, it is necessary to solve the problem of nonuniform thickness in the active layers of Gafchromic films. A double exposure technique using X-rays is performed in therapeutic radiology; it is difficult to use in a diagnostic examination because of the heel effect. Therefore, it is suggested that ultraviolet (UV) rays be substituted for X-rays. However, the appropriate UV wavelength is unknown. In this study, we aimed to determine which UV wavelengths are effective to expose Gafchromic XR-RV3 and XR-SP2. UV lamps with peak wavelengths of 245 nm, 310 nm, and 365 nm were used. The three UV wavelengths were used to irradiate Gafchromic XR-RV3 and XR-SP2 films for 60 min, and irradiation was repeated every 60 min for 600 min thereafter. Films were scanned after each irradiation period on a flatbed scanner. The images were split into their red-green-blue components, and red images were stored using ImageJ version 1.44o image analysis software. Regions of interest (ROI), 0.5 inches in diameter, were placed at the centers of the subtracted Gafchromic film images, and graphs of UV irradiation duration and mean pixel values were plotted. There were reactions to UV-A on both Gafchromic XR-RV3 and XR-SP2; those to UV-B were moderate. However, UV-C demonstrated few reactions with Gafchromic XR-RV3 and XR-SP2. From these results, irradiation with UV-A may be able to correct nonuniformity errors. Uniform UV-A irradiation of Gafchromic films with large areas is possible, and UV rays can be used as a substitute for X-rays in the double exposure technique.

Comparing three UV wavelengths for pre-exposing Gafchromic EBT2 and EBT3 films.

Gafchromic films are used for X-ray dose measurements during diagnostic examinations and have begun to be used for three-dimensional X-ray dose measurements using the high-resolution characteristics of Gafchromic films for computed tomography. However, the problem of unevenness in Gafchromic film active layers needs to be resolved. Double exposures using X-rays are performed during therapeutic radiology, although this is difficult for a diagnostic examination because of a heel effect. Thus, it has been suggested that ultraviolet (UV) radiation be used as a substitute for X-rays. However, the appropriate UV wavelength has not been determined. Thus, we conducted this study to decide an appropriate UV wavelength. UV peak wavelengths of 245 nm (UV-A), 310 nm (UV-B), and 365 nm (UV-C) were used to irradiate EBT2 and EBT3 films. Each UV wavelength was irradiated for 5, 15, 30, and 60 min, and irradiation was then repeated every 60 min up to 360 min. Gafchromic films were scanned after every irradiation using a flatbed scanner. Images were split into RGB images, and red images were analyzed using ImageJ, version 1.44, image analysis software. A region of interest (ROI) one-half inch in diameter was placed in the center of subtracted Gafchromic film images, and UV irradiation times were plotted against mean pixel values. There were reactions in the front and back of Gafchromic EBT3 and the back of Gafchromic EBT2 with UV-A and UV-B. However, UV-C resulted in some reactions in both sides of Gafchromic EBT2 and EBT3. The UV-A and UV-B wavelengths should be used.

Evaluation of GAFCHROMIC EBT2 dosimetry for the low dose range using a flat-bed scanner with the reflection mode.

Recently developed radiochromic films can easily be used to measure absorbed doses because they do not need development processing and indicate a density change that depends on the absorbed dose. However, in GAFCHROMIC EBT2 dosimetry (GAF-EBT2) as a radiochromic film, the precision of the measurement was compromised, because of non-uniformity problems caused by image acquisition using a flat-bed scanner with a transmission mode. The purpose of this study was to improve the precision of the measurement using a flat-bed scanner with a reflection mode at the low absorbed dose dynamic range of GAF-EBT2. The calibration curves of the absorbed dose versus the film density for GAF-EBT2 were provided. X-rays were exposed in the range between ~0 and 120 mGy in increments of about 12 mGy. The results of the method using a flat-bed scanner with the transmission mode were compared with those of the method using the same scanner with the reflection mode. The results should that the determination coefficients (r (2) ) for the straight-line approximation of the calibration curve using the reflection mode were higher than 0.99, and the gradient using the reflection mode was about twice that of the one using the transmission mode. The non-uniformity error that is produced by a flat-bed scanner with the transmission mode setting could be almost eliminated by converting from the transmission mode to the reflection mode. In light of these findings, the method using a flat-bed scanner with the reflection mode (only using uniform white paper) improved the precision of the measurement for the low absorbed dose range.

Improved detection of gastric cancer during screening by additional radiographs as judged necessary by the radiographer.

PURPOSE: The aim of this study was to determine whether additional radiographs, as judged necessary by the radiographer, improves cancer detection during gastric cancer screening. MATERIALS AND METHODS: We analyzed 144 gastric cancer cases among 137 744 individuals who underwent X-ray screening for gastric cancer. Radiographs were obtained by 17 radiographers at a screening center in Japan from April 2004 to March 2008. Additional radiographs were taken based on the radiographer's judgment in cases of suspected cancer. During double-blind reinterpretation of the cancer case radiographs by two radiologists, we determined the number of cancer cases that were detected by standard radiographs alone. We next determined the number of cancer cases detected using both standard radiographs and additional radiographs. RESULTS: Compared to the number of cancer cases detected with standard radiographs alone (120 cases detected, 24 cases undetected), the number of cancer cases detected with both standard and additional radiographs (137 cases detected, 7 cases undetected) significantly increased (17 cases; P < 0.001, McNemar test). CONCLUSION: We found that taking additional radiographs, when judged necessary by the radiographer during radiographic gastric cancer screening, improves cancer detection.

Film-reading ability of radiographers in detecting gastric cancer during screening using X-ray examination.

PURPOSE: The aim of this study was to evaluate the film-reading ability of radiographers in detecting gastric cancer during screening X-ray examinations. MATERIALS AND METHODS: A test set of 100 patients (50 negative and 50 positive; mean age 62 years, range 33-78 years) given a stomach X-ray examination were selected from those who underwent gastric cancer screening in Osaka, Japan, between 2000 and 2003. Eleven radiographers and four radiologists scored the test set on a five-point scale. A receiver operating characteristic (ROC) analysis was performed, and the area under the ROC curve (AUC) was defined as a measure of film-reading ability to detect cancer. RESULTS: No significant difference (two-tailed P = 0.962, Welch's t-test) was observed between averaged AUC values from radiographers (0.76, range 0.85-0.62) and radiologists (0.75, range 0.86-0.62). CONCLUSION: Film-reading ability of radiographers in detecting gastric cancer during screening X-ray examinations was not significantly different from that of radiologists. Our results suggest that radiographers can assist radiologists to detect gastric cancer during screening.

High resolution T2 weighted liver MR imaging using functional residual capacity breath-hold with a 1.0-Tesla scanner.

PURPOSE: During acquisition of rapid high resolution (HR) T2 weighted (T2W) liver magnetic resonance (MR) images using a 1.0-Tesla (T) scanner, the liver is segmented into odd and even sections that are acquired at two different times using the multi-breath-hold (MBH) strategy. Misalignment between the two breath-hold (B-H) images may result in the occurrence of a blind area and a decrease in diagnostic accuracy. Here, a functional residual capacity (FRC) B-H method was developed to overcome this problem. MATERIAL AND METHODS: Twenty-five volunteers were enrolled. The sagittal images were reconstructed from whole liver transverse images. When the B-H phases are different, misalignment may occur in the craniocaudal and anterior-posterior (AP) directions. In this study, misalignments of the abdominal wall were measured in the AP direction. The misalignment was compared between four B-H phases, maximum inspiration (MI), maximum expiration (ME), voluntary expiration (VE) and FRC using one-way repeated measures ANOVA. Differences between groups were compared using the t-test for multi-group comparisons. In addition, qualitative analysis of misalignment was performed between VE and FRC in 52 clinical patients and the chi(2) test was performed. RESULTS: The misalignment widths of FRC, ME, MI and VE B-Hs were 2.7+/-3.8, 6.4+/-7.4, 9.1+/-8.4 and 6.0+/-6.7 mm, respectively. Misalignment of the liver position using FRC was significantly smaller than for the other B-H methods (p<0.05). Significant differences between the VE B-H and FRC B-H were also observed in the qualitative analysis (p<0.05). CONCLUSION: The liver positions obtained when using FRC B-H were significantly more reproducible than when using the other B-H methods. The FRC B-H method resulted in a reduction in the blind area and an extension of the diagnostic area to the whole liver.

Functional residual capacity breath hold for subtraction image of dynamic liver MRI.

PURPOSE: In dynamic liver magnetic resonance imaging (MRI) studies, there are problems with misregistration when subtraction images are processed. For reduction of the misregistration, a functional residual capacity (FRC) phase breath-hold (FRC B-H) method was used. MATERIAL AND METHODS: Sixty patients (32 males and 28 females, aged 33-85 years, median age 69 years) were examined. The subjects were chronologically categorized into two groups: a voluntary expiratory (VE) B-H group and a FRC B-H group. The blood-flow phase images were classified as plain, arterial, portal and parenchymal phases. To evaluate the reproducibility of liver positions between VE B-H and FRC B-H in each phase (between Plain and Arterial, Arterial and Portal, Plain and Parenchymal), the misregistration areas were compared on the top of the liver. RESULTS: The misregistration area between Plain-Arterial, Arterial-Portal and Plain-Parenchymal in VE B-H was 731.0+/-1153.6, 1134.9+/-1357.2 and 628.4+/-844.5 (cm(2)), respectively. The misregistration area between each phase in FRC B-H was 386.4+/-874.9, 574.5+/-1086.1 and 279.8+/-551.2 (cm(2)), respectively. Using the Mann-Whitney U-test as quantitative analysis, the difference in misregistration areas between two groups was statistically significant (p<0.05). Differences in the qualitative analysis were also significant according to the chi(2) test (p<0.05). CONCLUSION: The liver positions with FRC B-H were markedly more reproducible than those with VE B-H. To improve the registration accuracy of subtraction dynamic liver MRI, the FRC B-H should be used.