Thin-slice 2D MR Imaging of the Shoulder Joint Using Denoising Deep Learning Reconstruction Provides Higher Image Quality Than 3D MR Imaging

PURPOSE: This study was conducted to evaluate whether thin-slice 2D fat-saturated proton density-weighted images of the shoulder joint in three imaging planes combined with parallel imaging, partial Fourier technique, and denoising approach with deep learning-based reconstruction (dDLR) are more useful than 3D fat-saturated proton density multi-planar voxel images. METHODS: Eighteen patients who underwent MRI of the shoulder joint at 3T were enrolled. The denoising effect of dDLR in 2D was evaluated using coefficient of variation (CV). Qualitative evaluation of anatomical structures, noise, and artifacts in 2D after dDLR and 3D was performed by two radiologists using a five-point Likert scale. All were analyzed statistically. Gwet's agreement coefficients were also calculated. RESULTS: The CV of 2D after dDLR was significantly lower than that before dDLR (P < 0.05). Both radiologists rated 2D higher than 3D for all anatomical structures and noise (P < 0.05), except for artifacts. Both Gwet's agreement coefficients of anatomical structures, noise, and artifacts in 2D and 3D produced nearly perfect agreement between the two radiologists. The evaluation of 2D tended to be more reproducible than 3D. CONCLUSION: 2D with parallel imaging, partial Fourier technique, and dDLR was proved to be superior to 3D for depicting shoulder joint structures with lower noise.

The Efficacy of CT Temporal Subtraction Images for Fibrodysplasia Ossificans Progressiva.

PURPOSE: To evaluate the usefulness of CT temporal subtraction (TS) images for detecting emerging or growing ectopic bone lesions in fibrodysplasia ossificans progressiva (FOP). MATERIALS AND METHODS: Four patients with FOP were retrospectively included in this study. TS images were produced by subtracting previously registered CT images from the current images. Two residents and two board-certified radiologists independently interpreted a pair of current and previous CT images for each subject with or without TS images. Changes in the visibility of the lesion, the usefulness of TS images for lesions with TS images, and the interpreter's confidence level in their interpretation of each scan were assessed on a semiquantitative 5-point scale (0-4). The Wilcoxon signed-rank test was used to compare the evaluated scores between datasets with and without TS images. RESULTS: The number of growing lesions tended to be larger than that of the emerging lesions in all cases. A higher sensitivity was found in residents and radiologists using TS compared to those not using TS. For all residents and radiologists, the dataset with TS tended to have more false-positive scans than the dataset without TS. All the interpreters recognized TS as useful, and confidence levels when using TS tended to be lower or the same as when not using TS for two residents and one radiologist. CONCLUSIONS: TS improved the sensitivity of all interpreters in detecting emerging or growing ectopic bone lesions in patients with FOP. TS could be applied further, including the areas of systematic bone disease.

Artificial intelligence-powered software detected more than half of the liver metastases overlooked by radiologists on contrast-enhanced CT.

PURPOSE: To evaluate the sensitivity of artificial intelligence (AI)-powered software in detecting liver metastases, especially those overlooked by radiologists. METHODS: Records of 746 patients diagnosed with liver metastases (November 2010-September 2017) were reviewed. Images from when radiologists first diagnosed liver metastases were reviewed, and prior contrast-enhanced CT (CECT) images were checked for availability. Two abdominal radiologists classified the lesions into overlooked lesions (all metastases missed by radiologists on prior CECT) and detected lesions (all metastases if any of them were correctly identified and invisible on prior CECT or those with no prior CECT). Finally, images from 137 patients were identified, 68 of which were classified as "overlooked cases." The same radiologists created the ground truth for these lesions and compared them with the software's output at 2-month intervals. The primary endpoint was the sensitivity in detecting all liver lesion types, liver metastases, and liver metastases overlooked by radiologists. RESULTS: The software successfully processed images from 135 patients. The per-lesion sensitivity for all liver lesion types, liver metastases, and liver metastases overlooked by radiologists was 70.1%, 70.8%, and 55.0%, respectively. The software detected liver metastases in 92.7% and 53.7% of patients in detected and overlooked cases, respectively. The average number of false positives was 0.48 per patient. CONCLUSION: The AI-powered software detected more than half of liver metastases overlooked by radiologists while maintaining a relatively low number of false positives. Our results suggest the potential of AI-powered software in reducing the frequency of overlooked liver metastases when used in conjunction with the radiologists' clinical interpretation.

Diagnostic advantage of thin slice 2D MRI and multiplanar reconstruction of the knee joint using deep learning based denoising approach.

The purpose of this study is to evaluate whether thin-slice high-resolution 2D fat-suppressed proton density-weighted image of the knee joint using denoising approach with deep learning-based reconstruction (dDLR) with MPR is more useful than 3D FS-PD multi planar voxel image. Twelve patients who underwent MRI of the knee at 3T and 13 knees were enrolled. Denoising effect was quantitatively evaluated by comparing the coefficient of variation (CV) before and after dDLR. For the qualitative assessment, two radiologists evaluated image quality, artifacts, anatomical structures, and abnormal findings using a 5-point Likert scale between 2D and 3D. All of them were statistically analyzed. Gwet's agreement coefficients were also calculated. For the scores of abnormal findings, we calculated the percentages of the cases with agreement with high confidence. The CV after dDLR was significantly lower than the one before dDLR (p < 0.05). As for image quality, artifacts and anatomical structure, no significant differences were found except for flow artifact (p < 0.05). The agreement was significantly higher in 2D than in 3D in abnormal findings (p < 0.05). In abnormal findings, the percentage with high confidence was higher in 2D than in 3D (p < 0.05). By applying dDLR to 2D, almost equivalent image quality to 3D could be obtained. Furthermore, abnormal findings could be depicted with greater confidence and consistency, indicating that 2D with dDLR can be a promising imaging method for the knee joint disease evaluation.

MR imaging features of mammary analogue secretory carcinoma and acinic cell carcinoma of the salivary gland: a preliminary report.

OBJECTIVES: To report MR imaging features of mammary analogue secretory carcinoma (MASC) and acinic cell carcinoma (AciCC) of the salivary gland based on the latest version of the World Health Organization (WHO) 2017 classification of head and neck tumours. METHODS: MR images in 4 patients with MASC and 4 with AciCC were reviewed for margin characteristics, the presence of pathological cervical nodes, the presence of a cystic component and interface between cystic and solid component, signal intensity of the cystic components on T1 weighted images, and signal intensity of the solid component on T1 and T2 weighted images. RESULTS: All the MASCs and AciCCs had well-defined boundaries, and 1 AciCC had pathological nodes. All 4 MASCs presented as predominantly cystic tumours with papillary projection of the solid component. All 4 AciCCs presented as solid tumours. The signal intensity of the cystic components on T1 weighted images was entirely hyperintense in 2, and partly hyperintense demonstrating fluid-fluid level in 2. In all the MASCs, the signal intensity of the solid components on T1 weighted images was intermediate. In the AciCCs, the signal intensity of the solid components on T1 weighted images was high in 2 tumours and intermediate in 2. The signal intensity of the solid components on T2 weighted images varied from low to high in both MSACs and AciCCs. CONCLUSIONS: All 4 MASCs had a large cystic component, including areas of high signal intensity on T1 weighted images. The solid component appeared as a papillary projection into the cystic component. All 4 AciCCs presented as solid tumours, 2 of which showed high signal intensity on T1 weighted images.

Subcutaneous fluid collection: An imaging marker for treatment response of infectious thoracolumbar spondylodiscitis.

PURPOSE: To evaluate prevalence of subcutaneous fluid collection (SFC) in infectious thoracolumbar spondylodiscitis (SD) compared with control patients and to investigate correlation between volume changes of SFC and treatment response of SD. MATERIALS AND METHODS: This retrospective study was approved by our institutional review board. From April 2011 to March 2012, 49 patients (24 SD and 25 non-SD patients) were enrolled. Prevalence of SFC was evaluated respectively for SD and non-SD patients using magnetic resonance imaging (MRI) on the sagittal short tau inversion recovery (STIR) imaging or fat-saturated T2-weighted imaging (T2WI), and compared. In SD patients with SFC, correlation was investigated between SFC volume on the 1st MRI and initial clinical status. The same analysis was conducted also for SFC volume changes from the 1st to 2nd or last MRI. RESULTS: SFC was found in 20 patients with SD (83.3%) and 3 non-SD patients (12%) with significant difference (p<.001). In 20 SD patients with SFC, 17 patients had follow-up MRI. For the 1st MRI, no significant correlation was found between volume of SFC and initial status of patients, including body weight, body mass index (BMI), white blood cell (WBC), and erythrocyte sedimentation rate (ESR). However, significant positive correlations were found between changes of C-reactive protein (CRP) and SFC volume from the 1st to 2nd as well as from the 1st to the last MRI (each p<.05). CONCLUSION: SD patients had significantly higher prevalence of SFC than non-SD patients. Volume changes of SFC had significant correlation with changes of CRP, which can be used as an imaging marker for treatment response of SD on MRI.

Estimation of proliferative potentiality of central neurocytoma: correlational analysis of minimum ADC and maximum SUV with MIB-1 labeling index.

BACKGROUND: Central neurocytoma was initially believed to be benign tumor type, although atypical cases with more aggressive behavior have been reported. Preoperative estimation for proliferating activity of central neurocytoma is one of the most important considerations for determining tumor management. PURPOSE: To investigate predictive values of image characteristics and quantitative measurements of minimum apparent diffusion coefficient (ADCmin) and maximum standardized uptake value (SUVmax) for proliferative activity of central neurocytoma measured by MIB-1 labeling index (LI). MATERIAL AND METHODS: Twelve cases of central neurocytoma including one recurrence from January 2001 to December 2011 were included. Preoperative scans were conducted in 11, nine, and five patients for computed tomography (CT), diffusion-weighted imaging (DWI), and fluorine-18-fluorodeoxyglucose positron emission tomography (FDG-PET), respectively, and ADCmin and SUVmax of the tumors were measured. Image characteristics were investigated using CT, T2-weighted (T2W) imaging and contrast-enhanced T1-weighted (T1W) imaging, and their differences were examined using the Fisher's exact test between cases with MIB-1 LI below and above 2%, which is recognized as typical and atypical central neurocytoma, respectively. Correlational analysis was conducted for ADCmin and SUVmax with MIB-1 LI. A P value <0.05 was considered significant. RESULTS: Morphological appearances had large variety, and there was no significant correlation with MIB-1 LI except a tendency that strong enhancement was observed in central neurocytomas with higher MIB-1 LI (P = 0.061). High linearity with MIB-1 LI was observed in ADCmin and SUVmax (r = -0.91 and 0.74, respectively), but only ADCmin was statistically significant (P = 0.0006). CONCLUSION: Central neurocytoma had a wide variety of image appearance, and assessment of proliferative potential was considered difficult only by morphological aspects. ADCmin was recognized as a potential marker for differentiation of atypical central neurocytomas from the typical ones.

Quantitative imaging values of CT, MR, and FDG-PET to differentiate pineal parenchymal tumors and germinomas: are they useful?

INTRODUCTION: Quantitative values of CT attenuation, apparent diffusion coefficient (ADC), and standardized uptake value (SUV) were investigated for differentiation between pineal parenchymal tumors (PPTs) and germinomas. Differences in age, sex, and calcification pattern were also evaluated. METHODS: Twenty-three patients with PPTs and germinomas in 20 years were retrospectively enrolled under the approval of the institutional review board. CT attenuation, ADC, and SUV (20, 13, and 10 patients, respectively) were statistically compared between the two tumors. Differences in sex and patterns of calcification ("exploded" or "engulfed") were also examined. Mean patient ages were compared among three groups of pineoblastoma, pineal parenchymal tumor of intermediate differentiation, (PPTID) and pineocytoma and germinoma. RESULTS: None of the quantitative values of CT attenuation, ADC, and SUV showed significant differences between PPTs and germinomas (p > .05). However, there was a significant difference in age (p < .05) among the three groups of pineoblastoma (mean age ± standard deviation 7.0 ± 8.7 years), PPTID, and pineocytoma (53.7 ± 11.4 years) and germinoma (19.1 ± 8.1 years). Sex also showed significant differences between PPTs and germinomas (p = .039). Exploded pattern of calcification was found in 9 of 11 PPT patients and engulfed pattern in 7 of 9 patients with germinomas. No reverse pattern was observed, and the patterns of calcification were considered highly specific of tumor types. CONCLUSIONS: None of the quantitative imaging values could differentiate PPTs from germinomas. Age, sex, and calcification patterns were confirmed useful in differentiating these tumors to some degree.

The iliac wing sign: an indicator of the presence of bone and/or soft-tissue injury of the pelvis and hips.

PURPOSE: To prospectively evaluate the feasibility of using the "iliac wing sign (IWS)" as an indicator of bone and/or soft-tissue injury of the pelvis and hips on magnetic resonance (MR) imaging. IWS means edema of the iliacus muscle attachment entering the iliac wing that is visualized as a linear high signal intensity on fat-suppressed T2-weighted MR images. METHODS: Consecutive 106 patients who complained of hip pain were enrolled in this study. We evaluated the correlation between IWS and bone and/or soft-tissue injury of the pelvis and hips using Fisher's exact test. Further, performance parameters of sensitivity, specificity, accuracy, the positive predictive value (PPV), and negative predictive value (NPV) of IWS were calculated. RESULTS: Thirty-eight of the 106 (36%) patients had bone and/or soft-tissue injury. Twenty-seven of these 38 (71%) patients with injury showed a positive IWS, while only 11 of 68 (16%) patients without injury showed a positive IWS (p<.0001). IWS, thus, yielded a sensitivity of 71%, specificity of 84%, accuracy of 79%, positive predictive value (PPV) of 71%, and negative predictive value (NPV) of 84%. CONCLUSION: In cases with a positive IWS, the careful interpretation of MR images is needed because injury presence is highly likely, as suggested by the relatively high sensitivity and PPV. IWS absence may mean a low probability of injury because of the high specificity and NPV.