Search

The study aimed to develop a deep neural network (DNN)-based noise reduction and image quality improvement by only using routine clinical scans and evaluate its performance in 3D high-resolution MRI.This retrospective study included T1-weighted magnetization-prepared rapid gradient-echo (MP-RAGE) images from 185 clinical scans: 135 for DNN training, 11 for DNN validation, 20 for qualitative evaluation, and 19 for quantitative evaluation. Additionally, 18 vessel wall imaging (VWI) data were included to evaluate generalization. In each scan of the DNN training set, two noise-independent images were generated from the k-space data, resulting in an input-label pair. 2.5D U-net architecture was utilized for the DNN model. Qualitative evaluation between conventional MP-RAGE and DNN-based MP-RAGE was performed by two radiologists in image quality, fine structure delineation, and lesion conspicuity. Quantitative evaluation was performed with full sampled data as a reference by measuring quantitative error metrics and volumetry at 7 different simulated noise levels. DNN application on VWI was evaluated by two radiologists in image quality.Our DNN-based MP-RAGE outperformed conventional MP-RAGE in all image quality parameters (average scores = 3.7 vs. 4.9, p0.001). In the quantitative evaluation, DNN showed better error metrics (p0.001) and comparable (p0.09) or better (p0.02) volumetry results than conventional MP-RAGE. DNN application to VWI also revealed improved image quality (3.5 vs. 4.6, p0.001).The proposed DNN model successfully denoises 3D MR image and improves its image quality by using routine clinical scans only.• Our deep learning framework successfully improved conventional 3D high-resolution MRI in all image quality parameters, fine structure delineation, and lesion conspicuity. • Compared to conventional MRI, the proposed deep neural network-based MRI revealed better quantitative error metrics and comparable or better volumetry results. • Deep neural network application to 3D MRI whose pulse sequences and parameters were different from the training data showed improvement in image quality, revealing the potential to generalize on various clinical MRI.

Purpose Nigrosome imaging using susceptibility-weighted imaging (SWI) and dopamine transporter imaging using 123I-2β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl)-nortropane (123I-FP-CIT) single-photon emission computerized tomography (SPECT) can evaluate Parkinsonism. Nigral hyperintensity from nigrosome-1 and striatal dopamine transporter uptake are reduced in Parkinsonism; however, quantification is only possible with SPECT. Here, we aimed to develop a deep-learning-based regressor model that can predict striatal 123I-FP-CIT uptake on nigrosome magnetic resonance imaging (MRI) as a biomarker for Parkinsonism. Methods Between February 2017 and December 2018, participants who underwent 3 T brain MRI including SWI and 123I-FP-CIT SPECT based on suspected Parkinsonism were included. Two neuroradiologists evaluated the nigral hyperintensity and annotated the centroids of nigrosome-1 structures. We used a convolutional neural network-based regression model to predict striatal specific binding ratios (SBRs) measured via SPECT using the cropped nigrosome images. The correlation between measured and predicted SBRs was evaluated. Results We included 367 participants (203 women (55.3%); age, 69.0 ± 9.2 [range, 39–88] years). Random data from 293 participants (80%) were used for training. In the test set (74 participants [20%]), the measured and predicted 123I-FP-CIT SBRs were significantly lower with the loss of nigral hyperintensity (2.31 ± 0.85 vs. 2.44 ± 0.90) than with intact nigral hyperintensity (4.16 ± 1.24 vs. 4.21 ± 1.35, P 123I-FP-CIT SBRs and the corresponding predicted values were significantly and positively correlated (ρc = 0.7443; 95% confidence interval, 0.6216–0.8314; P Conclusion A deep learning-based regressor model effectively predicted striatal 123I-FP-CIT SBRs based on nigrosome MRI with high correlation using manually-measured values, enabling nigrosome MRI as a biomarker for nigrostriatal dopaminergic degeneration in Parkinsonism.

Background Use of χ-separation imaging can provide surrogates for iron and myelin that relate closely to abnormal changes in multiple sclerosis (MS) lesions. Purpose To evaluate the appearances of MS and neuromyelitis optica spectrum disorder (NMOSD) brain lesions on χ-separation maps and explore their diagnostic value in differentiating the two diseases in comparison with previously reported diagnostic criteria. Materials and Methods This prospective study included individuals with MS or NMOSD who underwent χ-separation imaging from October 2017 to October 2020. Positive (χ

This study aimed to compare susceptibility map-weighted imaging (SMwI) using various MRI machines (three vendors) with N-3-fluoropropyl-2-β-carbomethoxy-3-β-(4-iodophe nyl)nortropane (We prospectively recruited 257 subjects, including 157 patients with neurodegenerative parkinsonism, 54 patients with non-neurodegenerative parkinsonism, and 46 healthy subjects from 10 hospitals between November 2019 and October 2020. All participants underwent both SMwI andThe inter-observer agreement was 0.924 and 0.942 per SN and participant, respectively. The diagnostic sensitivity of SMwI was 97.9% and 99.4% per SN and participant, respectively; its specificity was 95.9% and 95.2%, respectively, and its accuracy was 97.1% and 97.7%, respectively. There was no significant difference between the results of SMwI andThis study demonstrated that the high diagnostic performance of SMwI was maintained in a multi-centre setting with various MRI scanners, suggesting the generalisability of SMwI for determining nigrostriatal degeneration in patients with parkinsonism.• Susceptibility map-weighted imaging helps clinicians to predict nigrostriatal degeneration. • The protocol for susceptibility map-weighted imaging can be standardised across MRI vendors. • Susceptibility map-weighted imaging showed diagnostic performance comparable to that of dopamine transporter PET in a multi-centre setting with various MRI scanners.

Magnetic resonance imaging (MRI) can provide multiple contrast-weighted images using different pulse sequences and protocols. However, a long acquisition time of the images is a major challenge. To address this limitation, a new pulse sequence referred to as quad-contrast imaging is presented. The quad-contrast sequence enables the simultaneous acquisition of four contrast-weighted images (proton density (PD)-weighted, T2-weighted, PD-fluid attenuated inversion recovery (FLAIR), and T2-FLAIR), and the synthesis of T1-weighted images and T1- and T2-maps in a single scan. The scan time is less than 6 min and is further reduced to 2 min 50 s using a deep learning-based parallel imaging reconstruction. The natively acquired quad contrasts demonstrate high quality images, comparable to those from the conventional scans. The deep learning-based reconstruction successfully reconstructed highly accelerated data (acceleration factor 6), reporting smaller normalized root mean squared errors (NRMSEs) and higher structural similarities (SSIMs) than those from conventional generalized autocalibrating partially parallel acquisitions (GRAPPA)-reconstruction (mean NRMSE of 4.36% vs. 10.54% and mean SSIM of 0.990 vs. 0.953). In particular, the FLAIR contrast is natively acquired and does not suffer from lesion-like artifacts at the boundary of tissue and cerebrospinal fluid, differentiating the proposed method from synthetic imaging methods. The quad-contrast imaging method may have the potentials to be used in a clinical routine as a rapid diagnostic tool.

Background Dilated perivascular spaces (dPVS) are associated with aging and various disorders; however, the effect of age on dPVS burden in young populations and normative data have not been fully evaluated. Purpose To investigate the dPVS burden and provide normative data according to age in a healthy population, including children. Materials and Methods In this retrospective study, three-dimensional T2-weighted brain MRI scans from the Human Connectome Project data sets were used for visual grading (grade 0, 1, 2, 3, 4 for 0, 1-10, 11-20, 21-40, and40 dPVS on a single section of either hemispheric region) and automated volumetry of dPVS in basal ganglia (BGdPVS) and white matter (WMdPVS). Linear and nonlinear regression were performed to assess the association of dPVS volume with age. Optimal cutoff ages were determined with use of the maximized continuous-scale C-index. Participants were grouped by cutoff values. Linear regression was performed to assess the age-dPVS volume relationship in each age group. Normative data of dPVS visual grades were provided per age decade. Results A total of 1789 participants (mean age, 35 years; age range, 8-100 years; 1006 female participants) were evaluated. Age was related to dPVS volume in all regression models (

Parkinson disease is characterized by dopaminergic cell loss in the substantia nigra of the midbrain. There are various imaging markers for Parkinson disease. Recent advances in MRI have enabled elucidation of the underlying pathophysiologic changes in the nigral structure. This has contributed to accurate and early diagnosis and has improved disease progression monitoring. This article aims to review recent developments in nigral imaging for Parkinson disease and other parkinsonian syndromes, including nigrosome imaging, neuromelanin imaging, quantitative iron mapping, and diffusion-tensor imaging. In particular, this article examines nigrosome imaging using 7-T MRI and 3-T susceptibility-weighted imaging. Finally, this article discusses volumetry and its clinical importance related to symptom manifestation. This review will improve understanding of recent advancements in nigral imaging of Parkinson disease. Published under a CC BY 4.0 license.

Diffusion-weighted imaging (DWI) is proven useful to differentiate benign and malignant soft tissue tumors (STTs). Radiomics utilizing a vast array of extracted imaging features has a potential to uncover disease characteristics. We aim to assess radiomics using DWI can outperform the conventional DWI for STT differentiation. In 151 patients with 80 benign and 71 malignant tumors, ADCmean and ADCmin were measured on solid portion within the mass by two different readers. For radiomics approach, tumors were segmented and 100 original radiomic features were extracted on ADC map. Eight radiomics models were built with training set (n = 105), using combinations of 2 different algorithms—multivariate logistic regression (MLR) and random forest (RF)—and 4 different inputs: radiomics features (R), R + ADCmin (I), R + ADCmean (E), R + ADCmin and ADCmean (A). All models were validated with test set (n = 46), and AUCs of ADCmean, ADCmin, MLR-R, RF-R, MLR-I, RF-I, MLR-E, RF-E, MLR-A and RF-A models were 0.729, 0.753 0.698, 0.700, 0.773, 0.807, 0.762, 0.744, 0.773 and 0.807, respectively, without statistically significant difference. In conclusion, radiomics approach did not add diagnostic value to conventional ADC measurement for differentiating benign and malignant STTs.

Determination of preoperative soft tissue sarcoma (STS) margin is crucial for patient prognosis.To evaluate diagnostic performance of radiomics model using T2-weighted Dixon sequence for infiltration degree of STS margin.Retrospective.Seventy-two STS patients consisted of training (n = 58) and test (n = 14) sets.A 3.0 T; T2-weighted Dixon images.Pathologic result of marginal infiltration in STS (circumscribed margin; n = 27, group 1, focally infiltrative margin; n = 31, group 2-A, diffusely infiltrative margin; n = 14, group 2-B) was the reference standard. Radiomic volume and shape (VS) and other (T2) features were extracted from entire tumor volume and margin, respectively. Twelve radiomics models were generated using four combinations of classifier algorithms (R, SR, LR, LSR) and three different inputs (VS, T2, VS + T2 [VST2] features) to differentiate the three groups. Three radiologists (reader 1, 2, 3) analyzed the marginal infiltration with 6-scale confidence score.Area under the receiver operating characteristic curve (AUC) and concordance rate.Averaged AUCs of R, SR, LR, LSR models were 0.438, 0.466, 0.438, 0.466 using VS features, 0.596, 0.584, 0.814, 0.815 using T2 features, and 0.581, 0.587, 0.821, 0.821 using VST2 features, respectively. The LR and LSR models constructed with T2 or VST2 features showed higher AUC and concordance rate compared to radiologists' analysis (AUC; 0.730, 0.675, 0.706, concordance rate; 0.46, 0.43, 0.47 in reader 1, 2, 3).Radiomics model constructed with features from tumor margin on T2-weighted Dixon sequence is a promising method for differentiating infiltration degree of STS margin.4 TECHNICAL EFFICACY: Stage 2.

PURPOSE Synthetic magnetic resonance imaging (MRI) requires the acquisition of multicontrast images to estimate quantitative parameter maps, such as T1 , T2 , and proton density (PD). The study aims to develop a multicontrast reconstruction method based on joint parallel imaging (JPI) and joint deep learning (JDL) to enable further acceleration of synthetic MRI. METHODS The JPI and JDL methods are extended and combined to improve reconstruction for better-quality, synthesized images. JPI is performed as a first step to estimate the missing k-space lines, and JDL is then performed to correct and refine the previous estimate with a trained neural network. For the JDL architecture, the original variable splitting network (VS-Net) is modified and extended to form a joint variable splitting network (JVS-Net) to apply to multicontrast reconstructions. The proposed method is designed and tested for multidynamic multiecho (MDME) images with Cartesian uniform under-sampling using acceleration factors between 4 and 8. RESULTS It is demonstrated that the normalized root-mean-square error (nRMSE) is lower and the structural similarity index measure (SSIM) values are higher with the proposed method compared to both the JPI and JDL methods individually. The method also demonstrates the potential to produce a set of synthesized contrast-weighted images that closely resemble those from the fully sampled acquisition without erroneous artifacts. CONCLUSION Combining JPI and JDL enables the reconstruction of highly accelerated synthetic MRIs.

Objectives Despite its use in determining nigrostriatal degeneration, the lack of a consistent interpretation of nigrosome 1 susceptibility map-weighted imaging (SMwI) limits its generalized applicability. To implement and evaluate a diagnostic algorithm based on convolutional neural networks for interpreting nigrosome 1 SMwI for determining nigrostriatal degeneration in idiopathic Parkinson's disease (IPD). Methods In this retrospective study, we enrolled 267 IPD patients and 160 control subjects (125 patients with drug-induced parkinsonism and 35 healthy subjects) at our institute, and 24 IPD patients and 27 control subjects at three other institutes on approval of the local institutional review boards. Dopamine transporter imaging served as the reference standard for the presence or absence of abnormalities of nigrosome 1 on SMwI. Diagnostic performance was compared between visual assessment by an experienced neuroradiologist and the developed deep learning-based diagnostic algorithm in both internal and external datasets using a bootstrapping method with 10000 re-samples by the “pROC” package of R (version 1.16.2). Results The area under the receiver operating characteristics curve (AUC) (95% confidence interval [CI]) per participant by the bootstrap method was not significantly different between visual assessment and the deep learning-based algorithm (internal validation, .9622 [0.8912–1.0000] versus 0.9534 [0.8779-0.9956], P = .1511; external validation, 0.9367 [0.8843-0.9802] versus 0.9208 [0.8634-0.9693], P = .6267), indicative of a comparable performance to visual assessment. Conclusions Our deep learning-based algorithm for assessing abnormalities of nigrosome 1 on SMwI was found to have a comparable performance to that of an experienced neuroradiologist.

This study aimed to apply a radiomics approach to predict poor psychomotor development in preterm neonates using brain MRI. Prospectively enrolled preterm neonates underwent brain MRI near or at term-equivalent age and neurodevelopment was assessed at a corrected age of 12 months. Two radiologists visually assessed the degree of white matter injury. The radiomics analysis on white matter was performed using T1-weighted images (T1WI) and T2-weighted images (T2WI). A total of 1906 features were extracted from the images and the minimum redundancy maximum relevance algorithm was used to select features. A prediction model for the binary classification of the psychomotor developmental index was developed and eightfold cross-validation was performed. The diagnostic performance of the model was evaluated using the AUC with and without including significant clinical and DTI parameters. A total of 46 preterm neonates (median gestational age, 29 weeks; 26 males) underwent brain MRI (median corrected gestational age, 37 weeks). Thirteen of 46 (28.3%) neonates showed poor psychomotor outcomes. There was one neonate among 46 with moderate to severe white matter injury on visual assessment. For the radiomics analysis, twenty features were selected for each analysis. The AUCs of prediction models based on T1WI, T2WI, and both T1WI and T2WI were 0.925, 0.834, and 0.902. Including gestational age or DTI parameters did not improve the prediction performance of T1WI. A radiomics analysis of white matter using early T1WI or T2WI could predict poor psychomotor outcomes in preterm neonates. • Radiomics analysis on T1-weighted images of preterm neonates showed the highest diagnostic performance (AUC, 0.925) for predicting poor psychomotor outcomes. • In spite of 45 of 46 neonates having no significant white matter injury on visual assessment, the radiomics analysis of early brain MRI showed good diagnostic performance (sensitivity, 84.6%; specificity, 78.8%) for predicting poor psychomotor outcomes. • Radiomics analysis on early brain MRI can help to predict poor neurodevelopmental outcomes in preterm neonates.

BACKGROUND AND PURPOSE: Nigrostriatal dopaminergic function in patients with Parkinson disease can be assessed using (123)I-2β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl)-nortropan dopamine transporter ((123)I-FP-CIT) SPECT, and a good correlation has been demonstrated between nigral status on SWI and dopaminergic denervation on (123)I-FP-CIT SPECT. Here, we aim to correlate quantified dopamine transporter attenuation on (123)I-FP-CIT SPECT with nigrosome-1 status using susceptibility map-weighted imaging (SMWI). MATERIALS AND METHODS: Between May 2017 and January 2018, consecutive patients with idiopathic Parkinson disease (n = 109) and control participants (n = 29) who underwent (123)I-FP-CIT SPECT with concurrent 3T SWI were included. SMWI was generated from SWI. Two neuroradiologists evaluated nigral hyperintensity from nigrosome-1 on each side of the substantia nigra. Using consensus reading, we compared the (123)I-FP-CIT–specific binding ratio according to nigral hyperintensity status and the (123)I-FP-CIT specific binding ratio threshold to confirm the loss of nigral hyperintensity was determined using receiver operating characteristic curve analysis. RESULTS: The concordance rate between SMWI and (123)I-FP-CIT SPECT was 65.9%. The (123)I-FP-CIT–specific binding ratios in the striatum, caudate nucleus, and putamen were significantly lower when nigral hyperintensity in the ipsilateral substantia nigra was absent than when present (all, P

To evaluate the additional prognostic value of multiparametric MR-based radiomics in patients with glioblastoma when combined with conventional clinical and genetic prognostic factors. In this single-center study, patients diagnosed with glioblastoma between October 2007 and December 2019 were retrospectively screened and grouped into training and test sets with a 7:3 distribution. Segmentations of glioblastoma using multiparametric MRI were performed automatically via a convolutional-neural network. Prognostic factors in the clinical model included age, sex, type of surgery/post-operative treatment, and tumor location; those in the genetic model included statuses of isocitrate dehydrogenase-1 mutation and O-6-methylguanine-DNA-methyltransferase promoter methylation. Univariate and multivariate Cox proportional hazards analyses were performed for overall survival (OS) and progression-free survival (PFS). Integrated time-dependent area under the curve (iAUC) for survival was calculated and compared between prognostic models via the bootstrapping method (performances were validated with prediction error curves). Overall, 120 patients were included (training set, 85; test set, 35). The mean OS and PFS were 25.5 and 18.6 months, respectively. The prognostic performances of multivariate models improved when radiomics was added to the clinical model (iAUC: OS, 0.62 to 0.73; PFS, 0.58 to 0.66), genetic model (iAUC: OS, 0.59 to 0.67; PFS, 0.59 to 0.65), and combined model (iAUC: OS, 0.65 to 0.73; PFS, 0.62 to 0.67). In the test set, the combined model (clinical, genetic, and radiomics) demonstrated robust validation for risk prediction of OS and PFS. Radiomics increased the prognostic value when combined with conventional clinical and genetic prognostic models for OS and PFS in glioblastoma patients. • CNN-based automatic segmentation of glioblastoma on multiparametric MRI was useful in extracting radiomic features. • Patients with glioblastoma with high-risk radiomics scores had poor overall survival (hazards ratio 8.33, p < 0.001) and progression-free survival (hazards ratio 3.76, p < 0.001). • MR-based radiomics improved the survival prediction when combined with clinical and genetic factors (overall and progression-free survival iAUC from 0.65 to 0.73 and 0.62 to 0.67, respectively; both p < 0.001).

BACKGROUND AND PURPOSE: Human papillomavirus is a prognostic marker for oropharyngeal squamous cell carcinoma. We aimed to determine the value of CT-based radiomics for predicting the human papillomavirus status and overall survival in patients with oropharyngeal squamous cell carcinoma. MATERIALS AND METHODS: Eighty-six patients with oropharyngeal squamous cell carcinoma were retrospectively collected and grouped into training (n = 61) and test (n = 25) sets. For human papillomavirus status and overall survival prediction, radiomics features were selected via a random forest–based algorithm and Cox regression analysis, respectively. Relevant features were used to build multivariate Cox regression models and calculate the radiomics score. Human papillomavirus status and overall survival prediction were assessed via the area under the curve and concordance index, respectively. The models were validated in the test and The Cancer Imaging Archive cohorts (n = 78). RESULTS: For prediction of human papillomavirus status, radiomics features yielded areas under the curve of 0.865, 0.747, and 0.834 in the training, test, and validation sets, respectively. In the univariate Cox regression, the human papillomavirus status (positive: hazard ratio, 0.257; 95% CI, 0.09–0.7; P = .008), T-stage (≥III: hazard ratio, 3.66; 95% CI, 1.34–9.99; P = .011), and radiomics score (high-risk: hazard ratio, 3.72; 95% CI, 1.21–11.46; P = .022) were associated with overall survival. The addition of the radiomics score to the clinical Cox model increased the concordance index from 0.702 to 0.733 (P = .01). Validation yielded concordance indices of 0.866 and 0.720. CONCLUSIONS: CT-based radiomics may be useful in predicting human papillomavirus status and overall survival in patients with oropharyngeal squamous cell carcinoma.

Background and purpose Iron retained by activated microglia and macrophages in multiple sclerosis (MS) lesions may serve as a marker of innate immune system activation. Among several magnetic resonance imaging (MRI) methods, there has been recent interest in using quantitative susceptibility mapping (QSM) as a potential tool for assessing iron levels in the human brain. This study examined QSM findings in MS and neuromyelitis optica spectrum disorder (NMOSD) lesions obtained with 3-T MRI to assess imaging characteristics related to paramagnetic rims around brain lesions in MS and NMOSD. Methods This study included 32 MS and 21 seropositive NMOSD patients. MRI images were obtained using two 3-T MRI devices (Ingenia, Philips Healthcare; and Magnetom Verio, Siemens Healthineers) during routine diagnosis and treatment procedures. Multi and single echo gradient echo magnitude and phase images were obtained for QSM reconstruction. QSM images were used to characterize the detected lesions, and the findings were compared between MS and NMOSD. Results Totals of 71 and 35 MRI scans were performed during the study period in MS and NMOSD patients, respectively. In QSM images, paramagnetic rims were found in 26 (81.2%) MS patients and 1 (4.8%) NMOSD patient. Eight of the 22 MS patients and only 1 of the 10 NMOSD patients who underwent follow-up MRI showed new paramagnetic rims. The paramagnetic rim lesions appeared after enhancement or in new T2-weighted lesions without enhancement. Conclusions Paramagnetic rims might be a characteristic MRI finding for MS, and therefore they have potential as an imaging marker for differentially diagnosing MS from NMOSD using 3-T MRI.

Intracranial artery stenosis is an important public health concern internationally, due to it being one of the major causes of ischemic stroke. In this study, we aim to provide a computer-aided diagnosis algorithm capable of automatically distinguishing between Internal Carotid Artery (ICA) stenosis and normal to minimize the labor-intensiveness of stenosis detection. Using Time-of-Flight Magnetic Resonance Angiography (TOF-MRA), a novel deep learning detection model via 3D Squeeze and Excitation Residual Networks (SE-ResNet) is proposed. Pre-processing of TOF-MRA, data augmentation, training of 3D SE-ResNet, and testing using patch-based and patient-based methods with cross-validation is described. The proposed network using a database consisting of 50 normal cases (ICA-N) and 41 stenosis cases (ICA-S) with grade level of above 30% was evaluated. All 41 ICA-S cases were categorized according to the diameter (D_stenosis) of the artery at the site of the most severe stenosis by expert radiologists, whereas percent stenosis was measured by Warfarin-Aspirin Symptomatic Intracranial Disease (WASID) method. The proposed 3D SE-ResNet was further compared with more conventional networks including 3D ResNet and 3D VGG. The results showed the capability to detect stenosis achieving overall Area Under the Curve (AUC) and accuracies of 0.947 and 91.0% for patch-based and 0.884 and 81.0% for patient-based testing, respectively. In addition, the proposed 3D SE-ResNet achieved better performance against conventional 3D ResNet and 3D VGG with improvement rates of 0.053 and 0.095 for patch-based and 0.053 and 0.065 for patient-based testing in terms of AUC, respectively.

The purpose of this retrospective study was to investigate the association between ipsilateral recurrence of ductal carcinoma in situ (DCIS) and radiomics features from DCIS and contralateral normal breast on contrast enhanced breast MR imaging. A total of 163 patients with DCIS who underwent preoperative MR imaging between January 2010 and December 2014 were included (training cohort; n = 117, validation cohort; n = 46). Radiomics features were extracted from whole tumor volume of DCIS on early dynamic T1-subtraction images and from the contralateral normal breast on precontrast T1 and early dynamic T1-subtraction images. After feature selection, a Rad-score was established by LASSO Cox regression model. Performance of Rad-score was evaluated by the receiver operating characteristic (ROC) curve and Kaplan Meier curve with log rank test. The Rad-score was significantly associated with ipsilateral recurrence free survival (RFS). The low-risk group with a low Rad-score showed higher ipsilateral RFS than the high-risk group with a high Rad-score in both training and validation cohorts (p < 0.01). The Rad-score based on radiomics features from DCIS and contralateral normal breast on breast MR imaging showed the potential for prediction of ipsilateral RFS of DCIS.

Purpose This study aims to determine whether genetic factors affect the location of dilated perivascular spaces (dPVS) by comparing healthy young twins and non-twin (NT) siblings. Methods A total of 700 healthy young adult twins and NT siblings (138 monozygotic (MZ) twin pairs, 79 dizygotic (DZ) twin pairs, and 133 NT sibling pairs) were collected from the Human Connectome Project dataset. dPVS was automatically segmented and normalized to standard space. Then, spatial similarity indices (mean squared error [MSE], structural similarity [SSIM], and dice similarity [DS]) were calculated for dPVS in the basal ganglia (BGdPVS) and white matter (WMdPVS) between paired subjects before and after propensity score matching of dPVS volumes between groups. Within-pair correlations for the regional volumes of dVPS were also assessed using the intraclass correlation coefficient (ICC). Results The spatial similarity of dPVS was significantly higher in MZ twins (higher DS [median, 0.382 and 0.310] and SSIM [0.963 and 0.887] and lower MSE [0.005 and 0.005] for BGdPVS and WMdPVS, respectively) than DZ twins (DS [0.121 and 0.119], SSIM [0.941 and 0.868], and MSE [0.010 and 0.011]) and NT siblings (DS [0.106 and 0.097], SSIM [0.924 and 0.848], and MSE [0.016 and 0.017]). No significant difference was found between DZ twins and NT siblings. Similar results were found even after subjects were matched according to dPVS volume. Regional dPVS volumes were also more correlated within pairs in MZ twins than DZ twins and NT siblings. Conclusion Our results suggest that genetic factors affect the location of dPVS.

Background and Purpose Evaluation of dorsal nigral hyperintensity on MRI can help detect nigrostriatal degeneration. We aimed to compare the diagnostic performance between susceptibility map-weighted imaging (SMWI) and N-3-fluoropropyl-2-β-carbomethoxy-3-β-(4-iodophenyl) nortropane (18F-FP-CIT) positron emission tomography (PET) as an initial diagnostic tool of parkinsonism. Materials and methods This local ethics committee-approved retrospective study enrolled 223 patients with parkinsonism and 15 healthy subjects (mean age, 69.7 years; 135 females) who underwent both SMWI at 3T and 18F-FP-CIT PET. The diagnostic performances of the two tests for nigrostriatal degeneration were compared by evaluating whether the 90% confidence interval (CI) of the difference between the two tests was within the equivalence margin by using the DTComPair package of R. The concordance rate was tested by Cohen's kappa. Results The diagnostic sensitivities of SMWI and 18F-FP-CIT PET were 94.5% and 100% per SN and 100% and 100% per participant, respectively; their specificities were 95.3% and 86.7% per SN and 94.4% and 84.0% per participant, respectively. While the diagnostic sensitivity was comparable between the two tests for each SN and participant, the lower 90% CI of the differences in the specificity were −0.086 per SN and −0.104 per participant, indicating a higher diagnostic specificity of SMWI than that of 18F-FP-CIT PET. When excluding 20 participants with basal ganglia lesions, the two tests exhibited similar diagnostic performance and had excellent agreement (k = 0.899 per SN; k = 0.945 per participant). Conclusion For patients with parkinsonism, SMWI and 18F-FP-CIT PET exhibit similar diagnostic performance.

BACKGROUND FLAIR (fluid attenuated inversion recovery) imaging via synthetic MRI methods leads to artifacts in the brain, which can cause diagnostic limitations. The main sources of the artifacts are attributed to the partial volume effect and flow, which are difficult to correct by analytical modeling. In this study, a deep learning (DL)-based synthetic FLAIR method was developed, which does not require analytical modeling of the signal. PURPOSE To correct artifacts in synthetic FLAIR using a DL method. STUDY TYPE Retrospective. SUBJECTS A total of 80 subjects with clinical indications (60.6 ± 16.7 years, 38 males, 42 females) were divided into three groups: a training set (56 subjects, 62.1 ± 14.8 years, 25 males, 31 females), a validation set (1 subject, 62 years, male), and the testing set (23 subjects, 57.3 ± 20.4 years, 13 males, 10 females). FIELD STRENGTH/SEQUENCE 3 T MRI using a multiple-dynamic multiple-echo acquisition (MDME) sequence for synthetic MRI and a conventional FLAIR sequence. ASSESSMENT Normalized root mean square (NRMSE) and structural similarity (SSIM) were computed for uncorrected synthetic FLAIR and DL-corrected FLAIR. In addition, three neuroradiologists scored the three FLAIR datasets blindly, evaluating image quality and artifacts for sulci/periventricular and intraventricular/cistern space regions. STATISTICAL TESTS Pairwise Student's t-tests and a Wilcoxon test were performed. RESULTS For quantitative assessment, NRMSE improved from 4.2% to 2.9% (P

Gradient echo myelin water imaging (GRE-MWI) is an MRI technique to measure myelin concentration and involves the analysis of signal decay characteristics of the multi-echo gradient echo data. The method provides a myelin water fraction as a quantitative biomarker for myelin. In this work, a new sequence and post-processing methods were proposed to generate high quality GRE-MWI images at 3T and 7T. In order to capture the rapidly decaying myelin water signals, a bipolar readout GRE sequence was designed with "gradient pairing," compensating for the eddy current effects. The flip angle dependency from the multi-compartmental T1 effects was explored and avoided using a 2D multi-slice acquisition with a long TR. Additionally, the sequence was tested for the effects of inflow and magnetization transfer and demonstrated robustness to these error sources. Lastly, the temporal and spatial B0 inhomogeneity effects were mitigated by using the B0 navigator and field inhomogeneity corrections. Using the method, high-quality myelin water images were successfully generated for the in-vivo human brain at both field strengths. When the myelin water fraction at 3T and 7T were compared, they showed a good correlation (R2 ≥ 0.88; p

MR images based on phase contrast images have gained clinical interest as an in vivo tool for assessing anatomical and histological findings. The globus pallidus is an area of major iron metabolism and storage in the brain tissue. Calcium, another important metal in the body, is frequently deposited in the globus pallidus as well. Recently, we observed dense paramagnetic deposition with paradoxical calcifications in the globus pallidus and putamen. In this work, we explore detailed MR findings on these structures, and the histological source of the related findings using ex vivo CT and MR images. Ex vivo MR was obtained with a maximum 100 μm3 isotropic resolution using a 15.2 T MR system. 3D gradient echo images and quantitative susceptibility mapping were used because of their good sensitivity to metallic deposition, high signal-to-noise ratio, and excellent contrast to iron and calcium. We found dense paramagnetic deposition along the perforating arteries in the globus pallidus. This paramagnetic deposition was hyperdense on ex vivo CT scans. Histological studies confirmed this finding, and simultaneous deposition of iron and calcium, although more iron dominant, was observed along the vessel walls of the globus pallidus. This was an exclusive finding for the penetrating arteries of the globus pallidus. Thus, our results suggest that several strong and paradoxical paramagnetic sources at the globus pallidus can be associated with vascular degeneration.

Background It is important to diagnose sclerotic bone lesions in order to determine treatment strategy. Purpose To evaluate the diagnostic performance of a CT radiomics-based machine learning model for differentiating bone islands and osteoblastic bone metastases. Materials and Methods In this retrospective study, patients who underwent contrast-enhanced abdominal CT and were diagnosed with a bone island or osteoblastic metastasis between 2015 to 2019 at either of two different institutions were included: institution 1 for the training set and institution 2 for the external test set. Radiomics features were extracted. The random forest (RF) model was built using 10 selected features, and subsequent 10-fold cross-validation was performed. In the test phase, the RF model was tested with an external test set. Three radiologists reviewed the CT images for the test set. The sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated for the models and each of the three radiologists. The AUCs of the radiomics model and radiologists were compared. Results A total of 177 patients (89 with a bone island and 88 with metastasis; mean age, 66 years ± 12 [standard deviation]; 111 men) were in the training set, and 64 (23 with a bone island and 41 with metastasis; mean age, 69 years ± 14; 59 men) were in the test set. Radiomics features (n = 1218) were extracted. The average AUC of the RF model from 10-fold cross-validation was 0.89 (sensitivity, 85% [75 of 88 patients]; specificity, 82% [73 of 89 patients]; and accuracy, 84% [148 of 177 patients]). In the test set, the AUC of the trained RF model was 0.96 (sensitivity, 80% [33 of 41 patients]; specificity, 96% [22 of 23 patients]; and accuracy, 86% [55 of 64 patients]). The AUCs for the three readers were 0.95 (95% CI: 0.90, 1.00), 0.96 (95% CI: 0.90, 1.00), and 0.88 (95% CI: 0.80, 0.96). The AUC of radiomics model was higher than that of only reader 3 (0.96 vs 0.88, respectively; P = .03). Conclusion A CT radiomics-based random forest model was proven useful for differentiating bone islands from osteoblastic metastases and showed better diagnostic performance compared with an inexperienced radiologist. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Vannier in this issue.

Obtaining a histological fingerprint from the in-vivo brain has been a long-standing target of magnetic resonance imaging (MRI). In particular, non-invasive imaging of iron and myelin, which are involved in normal brain functions and are histopathological hallmarks in neurodegenerative diseases, has practical utilities in neuroscience and medicine. Here, we propose a biophysical model that describes the individual contribution of paramagnetic (e.g., iron) and diamagnetic (e.g., myelin) susceptibility sources to the frequency shift and transverse relaxation of MRI signals. Using this model, we develop a method, χ-separation, that generates the voxel-wise distributions of the two sources. The method is validated using computer simulation and phantom experiments, and applied to ex-vivo and in-vivo brains. The results delineate the well-known histological features of iron and myelin in the specimen, healthy volunteers, and multiple sclerosis patients. This new technology may serve as a practical tool for exploring the microstructural information of the brain.

Background The relationship between administration of macrocyclic gadolinium-based contrast agents and T1-weighted signal intensity (SI) change of the globus pallidus (GP) and dentate nucleus (DN) is, to the knowledge of the authors, not known. Purpose To determine if quantitative susceptibility mapping (QSM) can detect changes in magnetic susceptibility of the GP and DN after serial administration of macrocyclic gadobutrol in patients with primary brain tumors. Materials and Methods Patients diagnosed with primary brain tumors from August 2014 to February 2019 were eligible for this single-center retrospective study. Among 501 consecutive adult patients who were given at least one administration of gadobutrol, those who were previously administered an unknown or linear gadolinium-based contrast agent were excluded. Brain MRI scans with three-dimensional gradient-recalled-echo image phase data for QSM processing were reviewed. Regions of interest were drawn on the GP and DN on the basis of semiautomatic thresholding. Univariable generalized estimation equations were used to determine the associations between MRI measures (SI on T1-weighted images and magnetic susceptibility on QSM) and number of gadobutrol doses. Potential confounding factors were adjusted for in multivariable generalized estimating equation. Results Ninety patients (mean age, 51 years ± 17 [standard deviation]; 51 men) with 199 MRI scans were analyzed. In models adjusted for repeated observations between injections, the number of injections of gadobutrol was associated with the magnetic susceptibility of the GP (1.4 × 10

Background Automated measurement and classification models with objectivity and reproducibility are required for accurate evaluation of the breast cancer risk of fibroglandular tissue (FGT) and background parenchymal enhancement (BPE). Purpose To develop and evaluate a machine-learning algorithm for breast FGT segmentation and BPE classification. Study type Retrospective. Population A total of 794 patients with breast cancer, 594 patients assigned to the development set, and 200 patients to the test set. Field strength/sequence 3T and 1.5T; T2 -weighted, fat-saturated T1 -weighted (T1 W) with dynamic contrast enhancement (DCE). Assessment Manual segmentation was performed for the whole breast and FGT regions in the contralateral breast. The BPE region was determined by thresholding using the subtraction of the pre- and postcontrast T1 W images and the segmented FGT mask. Two radiologists independently assessed the categories of FGT and BPE. A deep-learning-based algorithm was designed to segment and measure the volume of whole breast and FGT and classify the grade of BPE. Statistical tests Dice similarity coefficients (DSC) and Spearman correlation analysis were used to compare the volumes from the manual and deep-learning-based segmentations. Kappa statistics were used for agreement analysis. Comparison of area under the receiver operating characteristic (ROC) curves (AUC) and F1 scores were calculated to evaluate the performance of BPE classification. Results The mean (±SD) DSC for manual and deep-learning segmentations was 0.85 ± 0.11. The correlation coefficient for FGT volume from manual- and deep-learning-based segmentations was 0.93. Overall accuracy of manual segmentation and deep-learning segmentation in BPE classification task was 66% and 67%, respectively. For binary categorization of BPE grade (minimal/mild vs. moderate/marked), overall accuracy increased to 91.5% in manual segmentation and 90.5% in deep-learning segmentation; the AUC was 0.93 in both methods. Data conclusion This deep-learning-based algorithm can provide reliable segmentation and classification results for BPE. Level of evidence 3 TECHNICAL EFFICACY STAGE: 2.

Acute appendicitis is one of the most common causes of abdominal emergencies. We investigated the feasibility of a neural-network-based diagnosis algorithm of appendicitis by using computed tomography (CT) for patients with acute abdominal pain visiting the emergency room (ER). A neural-network-based diagnostic algorithm of appendicitis was developed and validated using CT data from three institutions who visited the ER with abdominal pain and underwent abdominopelvic CT. For input data, 3D isotropic cubes including the appendix were manually extracted and labeled as appendicitis or a normal appendix. A 3D convolutional neural network (CNN) was trained to binary classification on the input. For model development and testing, 8-fold cross validation was conducted for internal validation and an ensemble model was used for external validation. Diagnostic performance was excellent in both the internal and external validation with an accuracy larger than 90%. The CNN-based diagnosis algorithm may be feasible in diagnosing acute appendicitis using the CT data of patients visiting the ER with acute abdominal pain.

PURPOSE To investigate the effect of the range of TE on multi-echo gradient echo-based myelin water fraction (MWF) mapping at 3 T. METHODS Myelin water fraction estimation was performed using 3 widely used models: magnitude multipool, magnitude 3-pool, and complex 3-pool model. Simulations were conducted using a hollow cylinder model with susceptibility anisotropy. The TE range was varied, and its effect on MWF estimation was observed. The same analysis was also performed with in vivo data from 4 healthy volunteers. RESULTS Different MWFs were obtained depending on the range of TE data acquired. In the simulations, the results of both magnitude models had a bias with respect to the true values: a maximum bias of up to 20% for the multipool model, and a maximum bias of up to 8% for the 3-pool model. This bias was reduced using the complex 3-pool model (maximum bias was under 2%). These effects were also found in the in vivo data. The MWF values using the complex model was most stable with respect to the TE range. CONCLUSION The results of the simulation and in vivo data suggest that MWF values fitted using magnitude models are more influenced by the TE range collected for fitting than using the complex model. The complex model is helpful in mitigating bias due to dependencies over the TE range selection.

Background Numerous studies have suggested that quantitative susceptibility mapping (QSM) and transverse relaxation rates ( R 2 * ) are useful to monitor neurological diseases. For clinical use of QSM and R 2 * , reproducibility is an important feature. However, respiration-induced local magnetic field variation makes artifacts in gradient echo-based images and reduces the reproducibility of QSM and R 2 * . Purpose To investigate the improvement of reproducibility of QSM and R 2 * after the correction of respiration-induced field variation, and to assess the effect of varying types of the region of interest (ROI) analysis on reproducibility. Study type Reproducibility study. Population Ten controls. Field strength/sequence 3T/multiecho gradient echo sequence. Assessment Intrascan reproducibility of QSM and R 2 * was investigated in ROIs before and after the respiration correction. Statistical tests Reproducibility was obtained by the square of voxel-wise correlation coefficients between scans. A paired t-test was performed for comparison between before and after the respiration correction and between QSM and R 2 * . Results Based on the ROI analysis, reproducibility increased after the respiration correction. Reproducibility in the white matter (11.89% increased in QSM and 23.38% in R 2 * , P = 0.009 and 0.024, respectively) and deep gray matter (5.50% increased in QSM and 13.96% in R 2 * , P = 0.024 and 0.019, respectively) increased significantly after the respiration correction. Reproducibility of R 2 * was higher than that of QSM in the whole brain and cortical gray matter, while QSM maps showed higher reproducibility than R 2 * in the white matter and deep gray matter. Data conclusion Respiration-induced error correction significantly improved reproducibility in QSM and R 2 * mapping. QSM and R 2 * mapping showed a different level of reproducibility depending on the types of ROI analysis. Level of evidence 4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018.