Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Background/Objectives: The clinical management of adolescent idiopathic scoliosis (AIS) is hindered by the inability to accurately predict curve progression. Although skeletal maturity and the initial Cobb angle are established predictors of progression, their combined predictive accuracy remains limited. This study aimed to develop a robust and interpretable artificial intelligence (AI) system using deep learning (DL) models to predict the progression of scoliosis using only standing frontal radiographs. Methods: We conducted a multicenter study involving 542 patients with AIS. After excluding 52 borderline progression cases (6-9° progression in the Cobb angle), 294 and 196 patients were assigned to progression (≥10° increase) and non-progression (≤5° increase) groups, respectively, considering a 2-year follow-up. Frontal whole spinal radiographs were preprocessed using histogram equalization and divided into two regions of interest (ROIs) (ROI 1, skull base-femoral head; ROI 2, C7-iliac crest). Six pretrained DL models, including convolutional neural networks (CNNs) and transformer-based models, were trained on the radiograph images. Gradient-weighted class activation mapping (Grad-CAM) was further performed for model interpretation. Results: Ensemble models outperformed individual ones, with the average ensemble model achieving area under the curve (AUC) values of 0.769 for ROI 1 and 0.755 for ROI 2. Grad-CAM revealed that the CNNs tended to focus on the local curve apex, whereas the transformer-based models demonstrated global attention across the spine, ribs, and pelvis. Models trained on ROI 2 performed comparably with respect to those using ROI 1, supporting the feasibility of image standardization without a loss of accuracy. Conclusions: This study establishes the clinical potential of transformer-based DL models for predicting the progression of scoliosis using only plain radiographs. Our multicenter approach, high AUC values, and interpretable architectures support the integration of AI into clinical decision-making for the early treatment of AIS.

DOI： 10.3390/jcm14207216

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

Predicting the risk of breast cancer recurrence is crucial for guiding therapeutic strategies, including enhanced surveillance and the consideration of additional treatment after surgery. In this study, we developed a deep convolutional neural network (DCNN) model to predict recurrence within six years after surgery using preoperative contrast-enhanced computed tomography (CECT) images, which are widely available and effective for detecting distant metastases. This retrospective study included preoperative CECT images from 133 patients with invasive ductal carcinoma. The images were classified into recurrence and no-recurrence groups using ResNet-101 and DenseNet-201. Classification performance was evaluated using the area under the receiver operating curve (AUC) with leave-one-patient-out cross-validation. At the optimal threshold, the classification accuracies for ResNet-101 and DenseNet-201 were 0.73 and 0.72, respectively. The median (interquartile range) AUC of DenseNet-201 (0.70 [0.69-0.72]) was statistically higher than that of ResNet-101 (0.68 [0.66-0.68]) (p < 0.05). These results suggest the potential of preoperative CECT-based DCNN models to predict breast cancer recurrence without the need for additional invasive procedures.

DOI： 10.1088/2057-1976/adeab5

PubMed

researchmap

Publishing type：Research paper (scientific journal)   Publisher：Japan Radioisotope Association  

DOI： 10.3769/radioisotopes.740103

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.3390/app15063034

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

researchmap

Authorship：Lead author,　Corresponding author   Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.11323/jjmp.45.1_3

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

PURPOSE: This study evaluates the use of deep learning techniques to automatically extract and delineate the aortic valve annulus region from contrast-enhanced cardiac CT images. Two approaches, namely, segmentation and object detection, were compared to determine their accuracy. MATERIALS AND METHODS: A dataset of 32 contrast-enhanced cardiac CT scans was analyzed. The segmentation approach utilized the DeepLabv3+ model, while the object detection approach employed YOLOv2. The dataset was augmented through rotation and scaling, and five-fold cross-validation was applied. The accuracy of both methods was evaluated using the Dice similarity coefficient (DSC), and their performance in estimating the aortic valve annulus area was compared. RESULTS: The object detection approach achieved a mean DSC of 0.809, significantly outperforming the segmentation approach, which had a mean DSC of 0.711. Object detection also demonstrated higher precision and recall, with fewer false positives and negatives. The aortic valve annulus area estimation had a mean error of 2.55 mm. CONCLUSIONS: Object detection showed superior performance in identifying the aortic valve annulus region, suggesting its potential for clinical application in cardiac imaging. The results highlight the promise of deep learning in improving the accuracy and efficiency of preoperative planning for cardiovascular interventions.

DOI： 10.3390/jcdd12010003

PubMed

researchmap

Language：Japanese   Publisher：鈴鹿 : 日本放射線技師教育学会  

researchmap

Other Link： https://ndlsearch.ndl.go.jp/books/R000000004-I033799028

Language：Japanese   Publishing type：Research paper (scientific journal)  

PURPOSE: The effects of scanning parameters such as CT system performance, CT bed geometry, and upper limb position on effective diameter (ED) and water equivalent diameter (WED) have not been assessed. The purpose of this study was to compare both ED and WED obtained with various CT systems with theoretical values and to assess their accuracy. METHODS: Jaszczak cylindrical phantom (Data Spectrum, Durham, NC, USA), NEMA IEC body phantom (AcroBio, Tokyo), and thoracic bone phantom were used in this study with and without upper limb phantom. The ED, WED, and size-specific dose estimate (SSDE) obtained using 8 types of CT systems were computed using radiation dose control software. RESULTS: The EDs had <5% error for all systems, but the error increased as the aspect ratio of the phantom increased. The accuracy of WED varied depending on the CT systems, with a maximum difference of 3.57 cm between systems. The influence of the upper limb depended on the shape of the bed of the CT systems, which affected the correlation between ED as well as WED and SSDE. CONCLUSION: Although the ED did not show any dependence on the CT system, the accuracy of WED for fusion CT was low. We found that there are issues in the management of scanning data, including the upper limb.

DOI： 10.6009/jjrt.2024-1511

PubMed

researchmap

Authorship：Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

PURPOSE: Currently, precise patient body weight (BW) at the time of diagnostic imaging cannot always be used for radiation dose management. Various methods have been explored to address this issue, including the application of deep learning to medical imaging and BW estimation using scan parameters. This study develops and evaluates machine learning-based BW prediction models using 11 features related to radiation dose obtained from computed tomography (CT) scans. METHODS: A dataset was obtained from 3996 patients who underwent positron emission tomography CT scans, and training and test sets were established. Dose metrics and descriptive data were automatically calculated from the CT images or obtained from Digital Imaging and Communications in Medicine metadata. Seven machine-learning models and three simple regression models were employed to predict BW using features such as effective diameter (ED), water equivalent diameter (WED), and mean milliampere-seconds. The mean absolute error (MAE) and correlation coefficient between the estimated BW and the actual BW obtained from each BW prediction model were calculated. RESULTS: Our results found that the highest accuracy was obtained using a light gradient-boosting machine model, which had an MAE of 1.99 kg and a strong positive correlation between estimated and actual BW (ρ = 0.972). The model demonstrated significant predictive power, with 73% of patients falling within a ±5% error range. WED emerged as the most relevant dose metric for BW estimation, followed by ED and sex. CONCLUSIONS: The proposed machine-learning approach is superior to existing methods, with high accuracy and applicability to radiation dose management. The model's reliance on universal dose metrics that are accessible through radiation dose management software enhances its practicality. In conclusion, this study presents a robust approach for BW estimation based on CT imaging that can potentially improve radiation dose management practices in clinical settings.

DOI： 10.1002/acm2.14467

PubMed

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)  

Cerebral computed tomography perfusion (CTP) imaging requires complete acquisition of contrast bolus inflow and washout in the brain parenchyma; however, time truncation undoubtedly occurs in clinical practice. To overcome this issue, we proposed a three-dimensional (two-dimensional + time) convolutional neural network (CNN)-based approach to predict missing CTP image frames at the end of the series from earlier acquired image frames. Moreover, we evaluated three strategies for predicting multiple time points. Seventy-two CTP scans with 89 frames and eight slices from a publicly available dataset were used to train and test the CNN models capable of predicting the last 10 image frames. The prediction strategies were single-shot prediction, recursive multi-step prediction, and direct-recursive hybrid prediction.Single-shot prediction predicted all frames simultaneously, while recursive multi-step prediction used prior predictions as input for subsequent steps, and direct-recursive hybrid prediction employed separate models for each step with prior predictions as input for the next step. The accuracies of the predicted image frames were evaluated in terms of image quality, bolus shape, and clinical perfusion parameters. We found that the image quality metrics were superior when multiple CTP images were predicted simultaneously rather than recursively. The bolus shape also showed the highest correlation (r = 0.990, p < 0.001) and the lowest variance (95% confidence interval, -453.26-445.53) in the single-shot prediction. For all perfusion parameters, the single-shot prediction had the smallest absolute differences from ground truth. Our proposed approach can potentially minimize time truncation errors and support the accurate quantification of ischemic stroke.

DOI： 10.1007/s12194-024-00818-6

PubMed

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.3390/app14145968

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.3390/app14093794

researchmap

Language：Japanese   Publisher：Japan Radioisotope Association  

We designed a patient–radiopharmaceutical matching system for preventing misadministration using smartphone, and developed a deep-learning model to identify radiopharmaceutical containers as an elemental technology for the system. As a result of the ResNet18 transfer learning and 10-fold cross-validation, this model achieved 100％ accuracy in classifying 15 different radiopharmaceutical containers. The feasibility of the proposed system was proven.

DOI： 10.3769/radioisotopes.73.69

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

OBJECTIVE: This study aimed to evaluate the correlation between the estimated body weight obtained from 2 easy-to-perform methods and the actual body weight at different computed tomography (CT) levels and determine the best reference site for estimating body weight. METHODS: A total of 862 patients from a public database of whole-body positron emission tomography/CT studies were retrospectively analyzed. Two methods for estimating body weight at 10 single-slice CT levels were evaluated: a linear regression model using total cross-sectional body area and a deep learning-based model. The accuracy of body weight estimation was evaluated using the mean absolute error (MAE), root mean square error (RMSE), and Spearman rank correlation coefficient (ρ). RESULTS: In the linear regression models, the estimated body weight at the T5 level correlated best with the actual body weight (MAE, 5.39 kg; RMSE, 7.01 kg; ρ = 0.912). The deep learning-based models showed the best accuracy at the L5 level (MAE, 6.72 kg; RMSE, 8.82 kg; ρ = 0.865). CONCLUSIONS: Although both methods were feasible for estimating body weight at different single-slice CT levels, the linear regression model using total cross-sectional body area at the T5 level as an input variable was the most favorable method for single-slice CT analysis for estimating body weight.

DOI： 10.1097/RCT.0000000000001587

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

This study aimed to evaluate the ability of deep learning reconstruction (DLR) compared to that of hybrid iterative reconstruction (IR) to depict small vessels on computed tomography (CT). DLR and two types of hybrid IRs were used for image reconstruction. The target vessels were the basilar artery (BA), superior cerebellar artery (SCA), anterior inferior cerebellar artery (AICA), and posterior inferior cerebellar artery (PICA). The peak value, ΔCT values defined as the difference between the peak value and background, and full width at half maximum (FWHM), were obtained from the profile curves. In all target vessels, the peak and ΔCT values of DLR were significantly higher than those of the two types of hybrid IR (p < 0.001). Compared to that associated with hybrid IR, the FWHM of DLR was significantly lower in the SCA (p < 0.001), AICA (p < 0.001), and PICA (p < 0.001). In conclusion, DLR has the potential to improve visualization of small vessels.

DOI： 10.1007/s12194-023-00749-8

PubMed

researchmap

Publishing type：Research paper (scientific journal)   Publisher：MDPI AG  

Convolutional neural networks (CNNs) in deep learning have input pixel limitations, which leads to lost information regarding microcalcification when mammography images are compressed. Segmenting images into patches retains the original resolution when inputting them into the CNN and allows for identifying the location of calcification. This study aimed to develop a mammographic calcification detection method using deep learning by classifying the presence of calcification in the breast. Using publicly available data, 212 mammograms from 81 women were segmented into 224 × 224-pixel patches, producing 15,049 patches. These were visually classified for calcification and divided into five subsets for training and evaluation using fivefold cross-validation, ensuring image consistency. ResNet18, ResNet50, and ResNet101 were used for training, each of which created a two-class calcification classifier. The ResNet18 classifier achieved an overall accuracy of 96.0%, mammogram accuracy of 95.8%, an area under the curve (AUC) of 0.96, and a processing time of 0.07 s. The results of ResNet50 indicated 96.4% overall accuracy, 96.3% mammogram accuracy, an AUC of 0.96, and a processing time of 0.14 s. The results of ResNet101 indicated 96.3% overall accuracy, 96.1% mammogram accuracy, an AUC of 0.96, and a processing time of 0.20 s. This developed method offers quick, accurate calcification classification and efficient visualization of calcification locations.

DOI： 10.3390/a16100483

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：MDPI AG  

Cardiac function indices must be calculated using tracing from short-axis images in cine-MRI. A 3D-CNN (convolutional neural network) that adds time series information to images can estimate cardiac function indices without tracing using images with known values and cardiac cycles as the input. Since the short-axis image depicts the left and right ventricles, it is unclear which motion feature is captured. This study aims to estimate the indices by learning the short-axis images and the known left and right ventricular ejection fractions and to confirm the accuracy and whether each index is captured as a feature. A total of 100 patients with publicly available short-axis cine images were used. The dataset was divided into training:test = 8:2, and a regression model was built by training with the 3D-ResNet50. Accuracy was assessed using a five-fold cross-validation. The correlation coefficient, MAE (mean absolute error), and RMSE (root mean squared error) were determined as indices of accuracy evaluation. The mean correlation coefficient of the left ventricular ejection fraction was 0.80, MAE was 9.41, and RMSE was 12.26. The mean correlation coefficient of the right ventricular ejection fraction was 0.56, MAE was 11.35, and RMSE was 14.95. The correlation coefficient was considerably higher for the left ventricular ejection fraction. Regression modeling using the 3D-CNN indicated that the left ventricular ejection fraction was estimated more accurately, and left ventricular systolic function was captured as a feature.

DOI： 10.3390/s23146580

PubMed

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

PURPOSE: Accurate body weight measurement is essential to promote computed tomography (CT) dose optimization; however, body weight cannot always be measured prior to CT examination, especially in the emergency setting. The aim of this study was to investigate whether deep learning-based body weight from chest CT scout images can be an alternative to actual body weight in CT radiation dose management. METHODS: Chest CT scout images and diagnostic images acquired for medical checkups were collected from 3601 patients. A deep learning model was developed to predict body weight from scout images. The correlation between actual and predicted body weight was analyzed. To validate the use of predicted body weight in radiation dose management, the volume CT dose index (CTDIvol ) and the dose-length product (DLP) were compared between the body weight subgroups based on actual and predicted body weight. Surrogate size-specific dose estimates (SSDEs) acquired from actual and predicted body weight were compared to the reference standard. RESULTS: The median actual and predicted body weight were 64.1 (interquartile range: 56.5-72.4) and 64.0 (56.3-72.2) kg, respectively. There was a strong correlation between actual and predicted body weight (ρ = 0.892, p < 0.001). The CTDIvol and DLP of the body weight subgroups were similar based on actual and predicted body weight (p < 0.001). Both surrogate SSDEs based on actual and predicted body weight were not significantly different from the reference standard (p = 0.447 and 0.410, respectively). CONCLUSION: Predicted body weight can be an alternative to actual body weight in managing dose metrics and simplifying SSDE calculation. Our proposed method can be useful for CT radiation dose management in adult patients with unknown body weight.

DOI： 10.1002/acm2.14080

PubMed

researchmap

Publishing type：Research paper (scientific journal)   Publisher：MDPI AG  

Although the widespread use of digital imaging has enabled real-time image display, images in chest X-ray examinations can be confirmed by the radiologist’s eyes. Considering the development of deep learning (DL) technology, its application will make it possible to immediately determine the need for a retake, which is expected to further improve examination throughput. In this study, we developed software for evaluating chest X-ray images to determine whether a repeat radiographic examination is necessary, based on the combined application of DL technologies, and evaluated its accuracy. The target population was 4809 chest images from a public database. Three classification models (CLMs) for lung field defects, obstacle shadows, and the location of obstacle shadows and a semantic segmentation model (SSM) for the lung field regions were developed using a fivefold cross validation. The CLM was evaluated using the overall accuracy in the confusion matrix, the SSM was evaluated using the mean intersection over union (mIoU), and the DL technology-combined software was evaluated using the total response time on this software (RT) per image for each model. The results of each CLM with respect to lung field defects, obstacle shadows, and obstacle shadow location were 89.8%, 91.7%, and 91.2%, respectively. The mIoU of the SSM was 0.920, and the software RT was 3.64 × 10−2 s. These results indicate that the software can immediately and accurately determine whether a chest image needs to be re-scanned.

DOI： 10.3390/app13116695

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

PURPOSE: Given the potential risk of motion artifacts, acquisition time reduction is desirable in pediatric 99m Tc-dimercaptosuccinic acid (DMSA) scintigraphy. The aim of this study was to evaluate the performance of predicted full-acquisition-time images from short-acquisition-time pediatric 99m Tc-DMSA planar images with only 1/5th acquisition time using deep learning in terms of image quality and quantitative renal uptake measurement accuracy. METHODS: One hundred and fifty-five cases that underwent pediatric 99m Tc-DMSA planar imaging as dynamic data for 10 min were retrospectively collected for the development of three deep learning models (DnCNN, Win5RB, and ResUnet), and the generation of full-time images from short-time images. We used the normalized mean squared error (NMSE), peak signal-to-noise ratio (PSNR), and structural similarity index metrics (SSIM) to evaluate the accuracy of the predicted full-time images. In addition, the renal uptake of 99m Tc-DMSA was calculated, and the difference in renal uptake from the reference full-time images was assessed using scatter plots with Pearson correlation and Bland-Altman plots. RESULTS: The predicted full-time images from the deep learning models showed a significant improvement in image quality compared to the short-time images with respect to the reference full-time images. In particular, the predicted full-time images obtained by ResUnet showed the lowest NMSE (0.4 [0.4-0.5] %) and the highest PSNR (55.4 [54.7-56.1] dB) and SSIM (0.997 [0.995-0.997]). For renal uptake, an extremely high correlation was achieved in all short-time and three predicted full-time images (R2  > 0.999 for all). The Bland-Altman plots showed the lowest bias (-0.10) of renal uptake in ResUnet, while short-time images showed the lowest variance (95% confidence interval: -0.14, 0.45) of renal uptake. CONCLUSIONS: Our proposed method is capable of producing images that are comparable to the original full-acquisition-time images, allowing for a reduction of acquisition time/injected dose in pediatric 99m Tc-DMSA planar imaging.

DOI： 10.1002/acm2.13978

PubMed

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Accurate body weights are not necessarily available in routine clinical practice. This study aimed to investigate whether body weight can be predicted from chest radiographs using deep learning. Deep-learning models with a convolutional neural network (CNN) were trained and tested using chest radiographs from 85,849 patients. The CNN models were evaluated by calculating the mean absolute error (MAE) and Spearman's rank correlation coefficient (ρ). The MAEs of the CNN models were 2.63 kg and 3.35 kg for female and male patients, respectively. The predicted body weight was significantly correlated with the actual body weight (ρ = 0.917, p < 0.001 for females; ρ = 0.915, p < 0.001 for males). The body weight was predicted using chest radiographs by applying deep learning. Our method is potentially useful for radiation dose management, determination of the contrast medium dose, and estimation of the specific absorption rate in patients with unknown body weights.

DOI： 10.1007/s12194-023-00697-3

PubMed

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

Consistent cross-sectional imaging is desirable to accurately detect lesions and facilitate follow-up in head computed tomography (CT). However, manual reformation causes image variations among technologists and requires additional time. We therefore developed a system that reformats head CT images at the orbitomeatal (OM) line and evaluated the system performance using real-world clinical data. Retrospective data were obtained for 681 consecutive patients who underwent non-contrast head CT. The datasets were randomly divided into one of three sets for training, validation, or testing. Four landmarks (bilateral eyes and external auditory canal) were detected with the trained You Look Only Once (YOLO)v5 model, and the head CT images were reformatted at the OM line. The precision, recall, and mean average precision at the intersection over union threshold of 0.5 were computed in the validation sets. The reformation quality in testing sets was evaluated by three radiological technologists on a qualitative 4-point scale. The precision, recall, and mean average precision of the trained YOLOv5 model for all categories were 0.688, 0.949, and 0.827, respectively. In our environment, the mean implementation time was 23.5 ± 2.4 s for each case. The qualitative evaluation in the testing sets showed that post-processed images of automatic reformation had clinically useful quality with scores 3 and 4 in 86.8%, 91.2%, and 94.1% for observers 1, 2, and 3, respectively. Our system demonstrated acceptable quality in reformatting the head CT images at the OM line using an object detection algorithm and was highly time efficient.

File： cover2022.jpg

DOI： 10.1007/s13246-022-01153-z

PubMed

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

Body weight is an indispensable parameter for determination of contrast medium dose, appropriate drug dosing, or management of radiation dose. However, we cannot always determine the accurate patient body weight at the time of computed tomography (CT) scanning, especially in emergency care. Time-efficient methods to estimate body weight with high accuracy before diagnostic CT scans currently do not exist. In this study, on the basis of 1831 chest and 519 abdominal CT scout images with the corresponding body weights, we developed and evaluated deep-learning models capable of automatically predicting body weight from CT scout images. In the model performance assessment, there were strong correlations between the actual and predicted body weights in both chest (ρ = 0.947, p < 0.001) and abdominal datasets (ρ = 0.869, p < 0.001). The mean absolute errors were 2.75 kg and 4.77 kg for the chest and abdominal datasets, respectively. Our proposed method with deep learning is useful for estimating body weights from CT scout images with clinically acceptable accuracy and potentially could be useful for determining the contrast medium dose and CT dose management in adult patients with unknown body weight.

DOI： 10.1038/s41598-021-95170-9

PubMed

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

PURPOSE: A variety of postprocessing algorithms for CT perfusion are available, with substantial differences in terms of quantitative maps. Although potential advantages of a Bayesian estimation algorithm are suggested, direct comparison with other algorithms in clinical settings remains scarce. We aimed to compare performance of a Bayesian estimation algorithm and singular value decomposition (SVD) algorithms for the assessment of acute ischemic stroke using an 80-detector row CT perfusion. METHODS: CT perfusion data of 36 patients with acute ischemic stroke were analyzed using the Vitrea implemented a standard SVD algorithm, a reformulated SVD algorithm and a Bayesian estimation algorithm. Correlations and statistical differences between affected and contralateral sides of quantitative parameters (cerebral blood volume [CBV], cerebral blood flow [CBF], mean transit time [MTT], time to peak [TTP] and delay) were analyzed. Agreement of the CT perfusion-estimated and the follow-up diffusion-weighted imaging-derived infarct volume were evaluated by nonparametric Passing-Bablok regression analysis. RESULTS: CBF and MTT of the Bayesian estimation algorithm were substantially different and showed a better correlation with the standard SVD algorithm (ρ = 0.78 and 0.80, p < 0.001) than with the reformulated SVD algorithm (ρ = 0.59 and 0.39, p < 0.001). There is no significant difference in MTT only when using the reformulated SVD algorithm (p = 0.217). Regarding the regression lines, the slope and intercept were nearly ideal with the Bayesian estimation algorithm (y = 2.42 x-6.51; ρ = 0.60, p < 0.001) in comparison with the SVD algorithms. CONCLUSIONS: The Bayesian estimation algorithm can lead to a better performance compared with the SVD algorithms in the assessment of acute ischemic stroke because of better delineation of abnormal perfusion areas and accurate estimation of infarct volume.

DOI： 10.1007/s11547-020-01316-6

PubMed

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

PURPOSE: To determine the optimal slice thickness of brain non-contrast computed tomography using a hybrid iterative reconstruction algorithm to identify hyperdense middle cerebral artery sign in patients with acute ischemic stroke. METHODS: We retrospectively enrolled 30 patients who had presented hyperdense middle cerebral artery sign and 30 patients who showed no acute ischemic change in acute magnetic resonance imaging. Reformatted axial images at an angle of the orbitomeatal line in slice thicknesses of 0.5, 1, 3, 5, and 7 mm were generated. Optimal slice thickness for identifying hyperdense middle cerebral artery sign was evaluated by a receiver operating characteristics curve analysis and area under the curve (AUC). RESULTS: The mean AUC value of 0.5-mm slice (0.921; 95% confidence interval (95% CI), 0.868 to 0.975) was significantly higher than those of 3-mm (0.791; 95% CI, 0.686 to 0.895; p = 0.041), 5-mm (0.691; 95% CI, 0.583 to 0.799, p < 0.001), and 7-mm (0.695; 95% CI, 0.593 to 0.797, p < 0.001) slices, whereas it was equivalent to that of 1-mm slice (0.901; 95% CI, 0.837 to 0.965, p = 0.751). CONCLUSION: Thin slice thickness of ≤ 1 mm has a better diagnostic performance for identifying hyperdense artery sign on brain non-contrast computed tomography with a hybrid iterative reconstruction algorithm in patients with acute ischemic stroke.

DOI： 10.1007/s10140-020-01864-4

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Elsevier {BV}  

PURPOSE: To assess the usefulness of effective diameter (Deff) for CT dose management of adult patients with unknown body weight. METHODS: A total of 642 adult patients whose height and weight had been measured before CT examination (chest CT using Aquilion Prime SP, 428 patients; chest CT using Biograph mCT, 100 patients; and abdominal CT using Aquilion Prime SP, 114 patients) were retrospectively examined between April 2018 and September 2019. The Deff was automatically calculated from the lateral diameter on a CT localizer radiograph by a dose management software (Radimetrics). In order to determine the correlation between body weight and Deff, we compared volume CT dose index and dose length product between patients with body weight between 50 and 70 kg and those with Deff equivalent to body weight between 50 and 70 kg. Correlation analysis was performed by Pearson's product-moment correlation, and statistical analyses were performed by using t-test. RESULTS: The correlation coefficient values between body weight and Deff were 0.920 for chest CT using Aquilion Prime SP, 0.929 for chest CT using Biograph mCT, and 0.805 for abdominal CT using Aquilion Prime SP. In both chest and abdominal CT scans, there were no significant differences in volume CT dose index and dose length product between patients with body weight between 50 and 70 kg and those with Deff equivalent to body weight between 50 and 70 kg. CONCLUSIONS: The Deff may be useful as a somatometric parameter for CT dose management of adult patients with unknown body weight.

DOI： 10.1016/j.ejrad.2020.109483

PubMed

researchmap

Authorship：Lead author,　Corresponding author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Ovid Technologies (Wolters Kluwer Health)  

PURPOSE: The purpose of this article was to verify the usefulness and feasibility of a single-phase scan for pulmonary artery/vein separation using a bolus-tracking technique and to evaluate the influence of patient characteristics on differentiation of computed tomography (CT) values between arteries and veins. MATERIAL AND METHODS: A total of 79 patients (60 male individuals and 19 female individuals, mean age 70 y) with suspected lung cancers or metastasis underwent contrast-enhanced chest CT and ultrasonic echocardiography. The CT values of the pulmonary arteries and veins were measured, and the difference in CT values was calculated. The relationships between the difference in CT values and age, weight, height, body surface area, body mass index, cardiac output, cardiac index, trigger time, trigger CT value, and pulmonary transit time were investigated using univariate linear regression analysis. RESULTS: The CT values were 352.8±87.3 HU and 494.6±76.5 HU for the pulmonary arteries and veins, respectively (P<0.001). A significant but weak correlation was seen between the difference in CT values and the height (r=0.24), trigger time (r=0.35), cardiac index (r=-0.25), and pulmonary transit time (r=0.53) (P<0.05). There was no significant correlation between the difference in CT values and the remaining values. CONCLUSION: The single-phase scan protocol using a bolus-tracking technique is feasible to differentiate CT values between pulmonary arteries and veins. The influence of patient characteristics on the differentiation of CT values lacks impact. Thus, the suggested protocol may be suitable independent of these factors after further validation.

DOI： 10.1097/rti.0000000000000455

PubMed

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Japanese Society of Radiological Technology  

DOI： 10.6009/jjrt.2020_jsrt_76.6.572

J-GLOBAL

researchmap

Other Link： https://search-tp.jamas.or.jp/index.php?module=Default&action=Link&pub_year=2020&ichushi_jid=J01192&link_issn=&doc_id=20200706440002&doc_link_id=1390848250120172544&url=https%3A%2F%2Fcir.nii.ac.jp%2Fcrid%2F1390848250120172544&type=CiNii&icon=https%3A%2F%2Fjk04.jamas.or.jp%2Ficon%2F00003_3.gif

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

The visual assessment of joint space narrowing (JSN) on radiographs of rheumatoid arthritis (RA) patients such as the Genant-modified Sharp score (GSS) is widely accepted but limited by its subjectivity and insufficient sensitivity. We developed a software application which can assess JSN quantitatively using a temporal subtraction technique for radiographs, in which the chronological change in JSN between two radiographs was defined as the joint space difference index (JSDI). The aim of this study is to prove the superiority of the software in terms of detecting fine radiographic progression in finger JSN over human observers. A micrometer measurement apparatus that can adjust arbitrary joint space width (JSW) in a phantom joint was developed to define true JSW. We compared the smallest detectable changes in JSW between the JSDI and visual assessment using phantom images. In a clinical study, 222 finger joints without interval score change on GSS in 15 RA patients were examined. We compared the JSDI between joints with and without synovial vascularity (SV) on power Doppler ultrasonography during the follow-up period. True JSW difference was correlated with JSDI for JSW differences ranging from 0.10 to 1.00 mm at increments of 0.10 mm (R2 = 0.986 and P < 0.001). Rheumatologists were difficult to detect JSW difference of 0.30 mm or less. The JSDI of finger joints with SV was significantly higher than those without SV (P = 0.030). The software can detect fine differences in JSW that are visually unrecognizable.

DOI： 10.1038/s41598-019-44747-6

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Wiley  

BACKGROUND: Synovitis, which is a hallmark of rheumatoid arthritis (RA), needs to be precisely quantified to determine the treatment plan. Time-intensity curve (TIC) shape analysis is an objective assessment method for characterizing the pixels as artery, inflamed synovium, or other tissues using dynamic contrast-enhanced MRI (DCE-MRI). PURPOSE/HYPOTHESIS: To assess the feasibility of our original arterial mask subtraction method (AMSM) with mutual information (MI) for quantification of synovitis in RA. STUDY TYPE: Prospective study. SUBJECTS: Ten RA patients (nine women and one man; mean age, 56.8 years; range, 38-67 years). FIELD STRENGTH/SEQUENCE: 3T/DCE-MRI. ASSESSMENT: After optimization of TIC shape analysis to the hand region, a combination of TIC shape analysis and AMSM was applied to synovial quantification. The MI between pre- and postcontrast images was utilized to determine the arterial mask phase objectively, which was compared with human subjective selection. The volume of objectively measured synovitis by software was compared with that of manual outlining by an experienced radiologist. Simple TIC shape analysis and TIC shape analysis combined with AMSM were compared in slices without synovitis according to subjective evaluation. STATISTICAL TESTS: Pearson's correlation coefficient, paired t-test and intraclass correlation coefficient (ICC). RESULTS: TIC shape analysis was successfully optimized in the hand region with a correlation coefficient of 0.725 (P < 0.01) with the results of manual assessment regarded as ground truth. Objective selection utilizing MI had substantial agreement (ICC = 0.734) with subjective selection. Correlation of synovial volumetry in combination with TIC shape analysis and AMSM with manual assessment was excellent (r = 0.922, P < 0.01). In addition, negative predictive ability in slices without synovitis pixels was significantly increased (P < 0.01). DATA CONCLUSIONS: The combination of TIC shape analysis and image subtraction reinforced with MI can accurately quantify synovitis of RA in the hand by eliminating arterial pixels. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.

DOI： 10.1002/jmri.25995

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：{SAGE} Publications  

Background Recent papers suggest that finger joints with positive synovial vascularity (SV) assessed by ultrasonography under clinical low disease activity (CLDA) in rheumatoid arthritis (RA) patients may cause joint space narrowing (JSN) progression. Purpose To investigate the performance of a computer-based method by directly comparing with the conventional scoring method in terms of the detectability of JSN progression in hand radiography of RA patients with CLDA. Material and Methods Fifteen RA patients (13 women, 2 men) with long-term sustained CLDA of >2 years were included. Radiological progression of finger joints was measured or scored using the computer-based method which can detect JSN progression between two radiographic images as the joint space difference index (JSDI), as well as the Genant-modified Sharp score (GSS). We also quantitatively assessed SV of these joints using ultrasonography. Results Out of 270 joints, we targeted 259 finger joints after excluding nine damaged joints (four ankylosis, three complete luxation, and two subluxation) and two improved joints according to the GSS results. The JSDI of finger joints with JSN progression was significantly higher than those without JSN progression ( P = 0.018). The JSDI of finger joints with ultrasonographic SV was significantly higher than those without ultrasonographic SV ( P = 0.004). Progression in JSDI showed stronger associations with ultrasonographic SV than progression in GSS (odds ratio [95% confidence interval]: 7.19 [3.37-15.36] versus 5.84 [2.76-12.33]). Conclusion The computer-based method was comparable to the conventional scoring method regarding the detectability of JSN progression in RA patients with CLDA.

DOI： 10.1177/0284185117721262

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

In previous studies, numerous approaches were proposed that assess knee cartilage volume quantitatively using 3D magnetic resonance (MR) imaging. However, the clinical use of these approaches is limited because 3D MR is prone to metal artifacts in postoperative cases. Our purpose in this study was to validate a method for knee cartilage volume quantification using conventional MR imaging in patients who underwent anterior cruciate ligament (ACL) reconstruction surgery. The study included 16 patients who underwent MR imaging before and 1 year after ACL reconstruction surgery. Knee cartilage volumes were measured by our computer-based method with the use of T1-weighted sagittal images. We classified the cartilage into eight regions and made comparisons between preoperative and postoperative cartilage volumes in each region. There was a significant difference between preoperative and postoperative cartilage volumes with regard to medial posterior weight-bearing, medial posterior, lateral posterior weight-bearing, and lateral posterior portions (p = 0.006, 0.023, 0.017 and 0.002, respectively). These results were consistent with the previous studies showing that knee cartilage loss occurs frequently in these portions due to an anterior subluxation of the tibia accompanied by ACL tear. With our method, knee cartilage volumes could be measured quantitatively with conventional MR imaging in patients who underwent ACL reconstruction surgery.

DOI： 10.1007/s12194-017-0415-4

PubMed

CiNii Article

researchmap

Other Link： https://search.jamas.or.jp/link/ui/2018321665

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

We have developed a refined computer-based method to detect joint space narrowing (JSN) progression with the joint space narrowing progression index (JSNPI) by superimposing sequential hand radiographs. The purpose of this study is to assess the validity of a computer-based method using images obtained from multiple institutions in rheumatoid arthritis (RA) patients. Sequential hand radiographs of 42 patients (37 females and 5 males) with RA from two institutions were analyzed by a computer-based method and visual scoring systems as a standard of reference. The JSNPI above the smallest detectable difference (SDD) defined JSN progression on the joint level. The sensitivity and specificity of the computer-based method for JSN progression was calculated using the SDD and a receiver operating characteristic (ROC) curve. Out of 314 metacarpophalangeal joints, 34 joints progressed based on the SDD, while 11 joints widened. Twenty-one joints progressed in the computer-based method, 11 joints in the scoring systems, and 13 joints in both methods. Based on the SDD, we found lower sensitivity and higher specificity with 54.2 and 92.8%, respectively. At the most discriminant cutoff point according to the ROC curve, the sensitivity and specificity was 70.8 and 81.7%, respectively. The proposed computer-based method provides quantitative measurement of JSN progression using sequential hand radiographs and may be a useful tool in follow-up assessment of joint damage in RA patients.

DOI： 10.1007/s10278-017-9970-9

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：British Institute of Radiology  

OBJECTIVE: To quantify wrist cartilage using contrast MRI and compare with the extent of adjacent synovitis and bone marrow edema (BME) in patients with rheumatoid arthritis (RA). METHODS: 18 patients with RA underwent post-contrast fat-suppressed T1weighted coronal imaging. Cartilage area at the centre of the scaphoid-capitate and radius-scaphoid joints was measured by in-house developed software. We defined cartilage as the pixels with signal intensity between two thresholds (lower: 0.4, 0.5 and 0.6 times the muscle signal, upper: 0.9, 1.0, 1.1, 1.2 and 1.3 times the muscle signal). We investigated the association of cartilage loss with synovitis and BME score derived from RA MRI scoring system. RESULTS: Cartilage area was correlated with BME score when thresholds were adequately set with lower threshold at 0.6 times the muscle signal and upper threshold at 1.2 times the muscle signal for both SC (rs=-0.469, p < 0.05) and RS (rs=-0.486, p < 0.05) joints, while it showed no significant correlation with synovitis score at any thresholds. CONCLUSION: Our software can accurately quantify cartilage in the wrist and BME associated with cartilage loss in patients with RA. Advances in knowledge: Our software can quantify cartilage using conventional MR images of the wrist. BME is associated with cartilage loss in RA patients.

DOI： 10.1259/bjr.20170167

PubMed

CiNii Article

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

The purpose of the study is to validate the semi-automated method using tomosynthesis images for the assessment of finger joint space narrowing (JSN) in patients with rheumatoid arthritis (RA), by using the semi-quantitative scoring method as the reference standard. Twenty patients (14 females and 6 males) with RA were included in this retrospective study. All patients underwent radiography and tomosynthesis of the bilateral hand and wrist. Two rheumatologists and a radiologist independently scored JSN with two modalities according to the Sharp/van der Heijde score. Two observers independently measured joint space width on tomosynthesis images using an in-house semi-automated method. More joints with JSN were revealed with tomosynthesis score (243 joints) and the semi-automated method (215 joints) than with radiography (120 joints), and the associations between tomosynthesis scores and radiography scores were demonstrated (P < 0.001). There was significant, negative correlation between measured joint space width and tomosynthesis scores with r = -0.606 (P < 0.001) in metacarpophalangeal joints and r = -0.518 (P < 0.001) in proximal interphalangeal joints. Inter-observer and intra-observer agreement of the semi-automated method using tomosynthesis images was in almost perfect agreement with intra-class correlation coefficient (ICC) values of 0.964 and 0.963, respectively. The semi-automated method using tomosynthesis images provided sensitive, quantitative, and reproducible measurement of finger joint space in patients with RA.

DOI： 10.1007/s10278-017-9949-6

PubMed

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

The joint space difference index (JSDI) is a newly developed radiographic index which can quantitatively assess joint space narrowing progression of rheumatoid arthritis (RA) patients by using an image subtraction method on a computer. The aim of this study was to investigate the reliability of this method by non-experts utilizing RA image evaluation. Four non-experts assessed JSDI for radiographic images of 510 metacarpophalangeal joints from 51 RA patients twice with an interval of more than 2 weeks. Two rheumatologists and one radiologist as well as the four non-experts examined the joints by using the Sharp-van der Heijde Scoring (SHS) method. The radiologist and four non-experts repeated the scoring with an interval of more than 2 weeks. We calculated intra-/inter-observer reliability using the intra-class correlation coefficients (ICC) for JSDI and SHS scoring, respectively. The intra-/inter-observer reliabilities for the computer-based method were almost perfect (inter-observer ICC, 0.966-0.983; intra-observer ICC, 0.954-0.996). Contrary to this, intra-/inter-observer reliability for SHS by experts was moderate to almost perfect (inter-observer ICC, 0.556-0.849; intra-observer ICC, 0.589-0.839). The results suggest that our computer-based method has high reliability to detect finger joint space narrowing progression in RA patients.

DOI： 10.1007/s00296-016-3588-y

PubMed

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：British Institute of Radiology  

OBJECTIVE: To investigate the validity of a computer-based method using temporal subtraction in carpal joints of patients with rheumatoid arthritis (RA), which can detect the difference in joint space between two images with the joint space difference index (JSDI). METHODS: The study consisted of 43 patients with RA (39 females and 4 males) who underwent radiography at baseline and at 1-year follow-up. The joint space narrowing (JSN) of carpal joints on bilateral hand radiographs was assessed by our computer-based method, using the Sharp/van der Heijde method as the standard of reference. We compared the JSDI of joints with JSN progression in the follow-up period with that of those without JSN progression. In addition, we examined whether there is a significant difference in JSDI in terms of laterality or topology of the joint. RESULTS: The JSDI of joints with JSN progression was significantly higher than that of those without JSN progression (Mann-Whitney U test, p < 0.001). There was no statistically significant difference in the JSDI between the left and right carpal joints, which was analysed for five different joints altogether and each joint separately (Mann-Whitney U test, p > 0.05). There was statistically significant difference in JSDI among different joints (Kruskal-Wallis test, p = 0.003). CONCLUSION: These results suggest that our computer-based method may be useful to recognize the JSN progression on radiographs of rheumatoid wrists. ADVANCES IN KNOWLEDGE: The computer-based temporal subtraction method can detect the JSN progression in the wrist, which is the single most commonly involved site in RA.

DOI： 10.1259/bjr.20150403

PubMed

CiNii Article

CiNii Books

researchmap

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media {LLC}  

Our computer-based method can detect the chronological change in joint space width between baseline and follow-up images as the joint space difference index (JSDI). The aim of this study was to verify the sensitivity and specificity of our computer-based method in assessment of joint space narrowing progression in rheumatoid patients. Twenty-seven patients (24 women and 3 men) with rheumatoid arthritis underwent radiography of the bilateral hand at baseline and at 1 year. The joint space narrowing (JSN) of a total of 252 metacarpophalangeal (MCP) joints and 229 carpal joints was assessed by our computer-based method, setting the Sharp/van der Heijde method as the gold standard. We constructed a receiver operating characteristic curve by using the Sharp/van der Heijde method as the gold standard and set the optimal cutoff on JSDI for MCP, carpal, and MCP/carpal joints. We then calculated the sensitivity and specificity for each cutoff in assessment of JSN progression. At the most discriminant cutoff, the sensitivity and specificity of the computer-based method for MCP joints was 78.6 versus 85.3 %, respectively (AUC = 0.837; P < 0.001). Carpal joints revealed a lower sensitivity and specificity with 64.7 and 86.8 % (AUC = 0.775; P < 0.001). Furthermore, the sensitivity and specificity for MCP/carpal joints was 71.0 versus 83.6 %, respectively (AUC = 0.778; P < 0.001). The computer-based method presented a reliable assessment of JSN progression with high sensitivity and specificity and may be useful in follow-up assessment of the joint damage in rheumatoid patients.

DOI： 10.1007/s00296-015-3349-3

PubMed

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会-北海道支部  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(株)インナービジョン  

J-GLOBAL

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本診療放射線技師会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

J-GLOBAL

researchmap

J-GLOBAL

researchmap

Language：Japanese   Publisher：北海道医学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本診療放射線技師会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：北海道医学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：北海道医学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本診療放射線技師会  

J-GLOBAL

researchmap

Language：English   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

Language：English   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

Language：English   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：北海道医学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：北海道医学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

Language：Japanese   Publisher：(公社)日本放射線技術学会  

J-GLOBAL

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

researchmap

Type：Academic society, research group, etc. 

researchmap

researchmap