T2 WI Research Articles

Imaging features of hemangioma in long tubular bones

BackgroundTo investigate the imaging features of hemangiomas in long tabular bones for better diagnosis.MethodsTwenty-four patients with long bone hemangiomas confirmed by pathology were enrolled. Nineteen patients had plain radiography, fourteen patients had computed tomography (CT) and eleven had magnetic resonance imaging (MRI). The hemangioma was divided into medullary [13], periosteal [6] and intracortical type [5].ResultsAmong 19 patients with plain radiography, eleven patients were medullary, three periosteal, and five intracortical. In the medullary type, the lesion was primarily osteolytic, including five cases with irregular and unclear rims and one lesion having osteosclerotic and unclear rims. In three patients with the periosteal type, the lesion had clear rims with involvement of the cortical bone in the form of bone defect, including two cases with local thickened bone periosteum and one case having expansile periosteum. Five intracortical hemangiomas had intracortical osteolytic lesions with clear margins. Among 14 patients with CT imaging, 8 cases were medullary, three periosteal, and three intracortical. Among 8 medullary hemangiomas, one had ground glass opacity, and seven had osteolytic, expansile lesions like soft tissue density with no calcification. In three periosteal cases, the lesion was osteolytic with thickened periosteum and narrowed medullary cavity. In three intracortical hemangiomas, the lesion was of even soft tissue density with no calcification. Among 11 patients with MRI imaging, seven were medullary, two periosteal, and two intracortical. Among 7 medullary lesions, six were of hypointense signal on T1WI and hyperintensesignal on T2 WI. In two periosteal cases, the periosteum was thickened, with one case being of equal signal, and the other having no signal. Two intracortical hemangiomas were both of slightly low signal on T1WI but hyperintense signal on T2WI.ConclusionsThe long bone hemangiomas had characteristic cystic honeycomb-like presentations in plain radiograph. CT and MRI imagings are helpful for diagnosis of hemangiomas in long bone.

Depiction of the Periosteum Using Ultrashort Echo Time Pulse Sequence with Three-Dimensional Cone Trajectory and Histologic Correlation in a Porcine Model.

ObjectiveTo evaluate the signal intensity of the periosteum using ultrashort echo time pulse sequence with three-dimensional cone trajectory (3D UTE) with or without fat suppression (FS) to distinguish from artifacts in porcine tibias.Materials and MethodsThe periosteum and overlying soft tissue of three porcine lower legs were partially peeled away from the tibial cortex. Another porcine tibia was prepared as three segments: with an intact periosteum outer and inner layer, with an intact periosteum inner layer, and without periosteum. Axial T1 weighted sequence (T1 WI) and 3D UTE (FS) were performed. Another porcine tibia without periosteum was prepared and subjected to 3D UTE (FS) and T1 WI twice, with positional changes. Two radiologists analyzed images to reach a consensus.ResultsThe three periosteal tissues that were partially peeled away from the cortex showed a high signal in 3D UTE (FS) and low signal on T1 WI. 3D UTE (FS) showed a high signal around the cortical surface with an intact outer and inner periosteum, and subtle high signals, mainly around the upper cortical surfaces with the inner layer of the periosteum and without periosteum. T1 WI showed no signal around the cortical surfaces, regardless of the periosteum state. The porcine tibia without periosteum showed changes in the high signal area around the cortical surface as the position changed in 3D UTE (FS). No signal was detected around the cortical surface in T1 WI, regardless of the position change.ConclusionThe periosteum showed a high signal in 3D UTE and 3D UTE FS that overlapped with artifacts around the cortical bone.

Post-operative changes following endoscopic third ventriculostomy by MR imaging.

Background: Endoscopic Third Ventriculostomy (ETV) has increased in popularity in recent years as an acceptable and effective treatment for many children with hydrocephalus. Purpose: To describe the post-operative findings after ETV with clinical correlation and to see if changes in ventricular size existed after ETV are correlated to clinical result. Methods and Materials: A total of 15 pediatric patients underwent ETV at our institution for hydrocephalus during the period of 2019 and 2020. MRI was routinely performed within the first week after surgery and its 3-month follow up. Post-operative clinical assessment was conducted by the neurosurgery team in the early post-operative period and at three and six months if the patient remained well. Results: The overall success rate and failure rate to restore CSF circulation was 73.3% and 26.7%, respectively. It demonstrated the strong association between clinical success and radiographic findings of ventriculostomy success by the presence of signal void within the expected site of third ventriculostoma on T2 weighted sequence (P=0.0088). Also, There was a statistically significant association between late clinical outcome and CSF collection with a signal void seen within it on T2 WI (P<0.001). Prepontine adhesions were seen in all the four failed ETVs. No apparent correlation between the magnitude of the change in ventricular size by Evans’ index even shortly after surgery or after three months and the clinical outcome Conclusion: The MRI study especially T2 weighted images can demonstrate successful ETV procedure by the presence of signal void at the ventriculostomy site, prepontine adhesions and a subcutaneous CSF collection with a signal void seen within it.

Alteration of single-subject gray matter networks in major depressed patients with suicidality.

While regional brain alterations and functional connectivity in depressed suicidal patients have previously been reported, knowledge about gray matter (GM) structural networks is limited. The aim of this study was to explore the GM of depressed suicidal brains from the single-subject structural network level. This was a cross-sectional study, in which 50 healthy controls (HC, 31 ± 9 years), 50 major depressed patients without suicidality (NSD, 29 ± 10 years), and 50 major depressed patients with suicidality (SU, 29 ± 12 years) were enrolled. T1 -weighted images (T1 WI) were acquired with three-dimensional-magnetization prepared rapid gradient echo sequence in 3.0 T magnetic resonance. The analysis was performed using the automated Computational Anatomy Toolbox (CAT12) within Statistical Parametric Mapping while running MATLAB. The T1 images were segmented into GM, white matter, and cerebrospinal fluid. Then single-subject structural networks were constructed based on the morphological similarity of GM regions. Global network topological properties, including clustering coefficient (Cp ), characterpath length (Lp ), normalized clustering coefficient (γ), normalized characteristic path length (λ), small-worldness (σ), global efficiency (Eglob ), local efficiency (Eloc ), and nodal network topological properties, including nodal efficiency, degree, and betweenness centrality, were measured using graph theory analysis. Statistical tests performed were analysis of variance, Pearson correlation analysis, and multiple linear regression analysis. Decreased Eglob and increased shortest Lp were observed in SU group compared to HC and NSD groups (p < 0.05). The NSD and SU groups had an increased λ and decreased Eloc compared to the HC group (p < 0.05). Altered nodal efficiency was found in the fronto-striatum-limbic-thalamic circuit in the SU group compared with the HC and NSD groups (all p < 0.05). The GM network in the SU group showed decreased segregation and weaker integration, that is weaker small-worldness, compared to the NSD and HC groups. Abnormal nodal efficiency was found in the fronto-striatum-limbic-thalamic circuit in suicidal brains. This study provides new evidence for therapeutic targets for patients with depression and suicidality. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 3.

Radiomics Analysis Based on Multiparametric MRI for Predicting Early Recurrence in Hepatocellular Carcinoma After Partial Hepatectomy.

Preoperative prediction of early recurrence (ER) of hepatocellular carcinoma (HCC) plays a critical role in individualized risk stratification and further treatment guidance. To investigate the role of radiomics analysis based on multiparametric MRI (mpMRI) for predicting ER in HCC after partial hepatectomy. Retrospective. In all, 113 HCC patients (ER, n = 58 vs. non-ER, n = 55), divided into training (n = 78) and validation (n = 35) cohorts. 1.5T or 3.0T, gradient-recalled-echo in-phase T1 -weighted imaging (I-T1 WI) and opposed-phase T1 WI (O-T1 WI), fast spin-echo T2 -weighted imaging (T2 WI), spin-echo planar diffusion-weighted imaging (DWI), and gradient-recalled-echo contrast-enhanced MRI (CE-MRI). In all, 1146 radiomics features were extracted from each image sequence, and radiomics models based on each sequence and their combination were established via multivariate logistic regression analysis. The clinicopathologic-radiologic (CPR) model and the combined model integrating the radiomics score with the CPR risk factors were constructed. A nomogram based on the combined model was established. Receiver operating characteristic (ROC) curve analysis was used to evaluate the discriminative performance of each model. The potential clinical usefulness was evaluated by decision curve analysis (DCA). The radiomics model based on I-T1 WI, O-T1 WI, T2 WI, and CE-MRI sequences presented the best performance among all radiomics models with an area under the ROC curve (AUC) of 0.771 (95% confidence interval (CI): 0.598-0.894) in the validation cohort. The combined nomogram (AUC: 0.873; 95% CI: 0.756-0.989) outperformed the radiomics model and the CPR model (AUC: 0.742; 95% CI: 0.577-0.907). DCA demonstrated that the combined nomogram was clinically useful. The mpMRI-based radiomics analysis has potential to predict ER of HCC patients after hepatectomy, which could enhance risk stratification and provide support for individualized treatment planning. 4. Stage 4.

Biological non-ablative repair of lumbar discs by transforaminal intradiscal laser irradiation: MRI quantitative analysis of the effects-preliminary report.

Degenerative disc disease is a significant reason for low back pain. Low-level laser irradiation (LLLI) of cartilage results in its reshaping and combines with regenerative reaction. A certain pattern of lumbar disc irradiation induces healing reaction and formation of new cartilage. Quantitative MRI analysis of regenerative response of the cartilage is the subject of this investigation. Fifty-one lumbar discs of 28 patients with discogenic low back pain underwent irradiation with 1.56-μm Er fiber laser (1.2W). Quantitative MRI analysis is performed in STIR regime within 0.93-14.80months. Signal intensity is estimated from irradiated discs and control measured from adjacent non-irradiated discs and vertebral bones. T2 WI follow-up is performed within a long period (up to 5years) in selected cases. The mean value of MRI signal intensity from the irradiated discs increased by 14% (p <<< 0.001). The control bone measurement revealed no difference in signal intensity (p = 0.83). The adjacent non-irradiated discs slightly increased their signal (p < 0.05). T2 WI follow-up within 5years revealed a steady increase of the signal and the irradiated discs healing. LLLI of degenerated intervertebral discs by 1.56-μm Er fiber laser produces increase of MRI disc signal within the first year after treatment that confirms regenerative response of the disc and could lay in the basis of clinical improvement. Further assessment on the effect is mandatory.

Hypervascular pancreatic metastasis: an easy imaging diagnostic?

Learning objectives. To review the computed tomograpy (CT) and magnetic resonance imaging (MRI) features of hypervascular pancreatic metastases (HPM). To discuss the main differential diagnosis of HPM. Materials and method. We retrospectively analyzed images from the CT and MRI evaluations of six patients who were diagnosed with HPM by morphopathological examination in the last ten years. Results. In our study, hypervascular pancreatic metastases corresponded to renal cell carcinoma, breast carcinoma and uterine leiomyosarcoma. In non-enhanced CT (NECT), all lesions were isodense or slightly hypodense. After i.v. injection of contrast medium (CM), hypervascular pancreatic metastases have displayed an early arterial phase enhancement, followed by rapid washout in the portal and delayed phases images. Contrast-enhanced CT showed for pancreatic metastases from uterine leio­myo­sarcoma a gradual enhancement. The common ap­pea­rance in MRI for HPM was hypointense in T1 weighted image (wi), heterogeneous aspect in T2 wi or mo­de­ra­tely hyperintense signal with restricted diffusion. After Gd-based CM i.v. injection, HPM showed the same dy­namic as in CT evaluation. Conclusions. HPM may create a diagnostic challenge, therefore the pathological exa­mi­na­tion remains essential and represents the gold standard for the final diagnosis.

Optimized MRI Assessment for Clinically Significant Prostate Cancer: A STARD-Compliant Two-Center Study.

There is a requirement for a personalized strategy to make MRI more accessible to men with suspicion of clinically significant prostate cancer (CSPC). To evaluate an optimized (Op)-MRI compared with biparametric (Bp)-MRI and multiparametric (Mp)-MRI for the diagnosis of CSPC. Two-center, retrospective. A total of 346 patients from center 1 and 292 patients from center 2. 3.0T scanners, T2 -weighted imaging (T2 WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE) imaging. Four radiologists interpreted the Bp-MRI (T2 WI and DWI) and Mp-MRI (T2 WI, DWI, and DCE) independently according to the Prostate Imaging Reporting and Data System (PI-RADS). For Op-MRI, two radiologists used an adjusted decision rule on Bp-MRI-assessed PI-RADS 3 lesions by determining early enhancement of DCE. Pathologies at biopsy and/or prostatectomy specimens were used as standard references. Performance was assessed using receiver operating characteristic (ROC) curves. Kappa statistics were used to assess interobserver variability. Interreader agreement was excellent for all three MRI assessments (all κ values >0.80). Op-MRI had comparable sensitivity (senior/junior: 90.9% [261/287] / 91.6% [263/287]) and higher specificity (78.1% [274/351] /74.4% [261/351]) compared with Mp-MRI (sensitivity: 92.3% [265/287] / 93.7% [269/287]; specificity: 67.8% [238/351] / 68.1% [239/351]) and Bp-MRI (sensitivity: 91.6% [263/287] / 93.4% [268/287]; specificity: 71.2% [250/351] / 70.1% [246/351]) for the diagnosis of CSPC. Compared to Mp-MRI, Op-MRI spared biopsy in 80.7% (515/638) of DCE scans with similar performance accuracy. Compared to Bp-MRI, Op-MRI downgraded 25.2% (31/123) of lesions at a cost of missing 6.5% (3/46) of malignancies, and upgraded 45.5% (56/123) of lesions with a positive predictive value of 62.5% (35/56) in 123 equivocal findings. The Op-MRI, using an adjusted PI-RADS decision rule, did not compromise diagnostic accuracy with sparing biopsy in 80.7% of DCE scans compared to Mp-MRI, and outperformed Bp-MRI by regrading PI-RADS lesions. 4 TECHNICAL EFFICACY STAGE: 2.

MRI Texture-Based Models for Predicting Mitotic Index and Risk Classification of Gastrointestinal Stromal Tumors.

Treatment regimens and prognoses of gastrointestinal stromal tumors (GIST) are quite different for tumors in different risk categories. Accurate preoperative grading of tumors is important for avoiding under- or overtreatment. To develop and validate an MRI texture-based model to predict the mitotic index and its risk classification. Retrospective. Ninety-one patients with histologically-confirmed GIST; 64 patients in a training cohort, and 27 patients in a test cohort. T2 -weighted imaging (T2 WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced three-dimensional volumetric interpolated breath-hold examination (3D-VIBE) at 1.5T. GIST images were manually segmented by two independent radiologists using ITK-SNAP software and MRI features were extracted using Pyradiomics. Two pathologists reviewed the tissue specimens of the tumors to identify the mitotic index and risk classification in consensus. The least absolute shrinkage and selection operator (LASSO) regression method was used to select texture features. A logistic regression model was established based on the radiomic score (radscore), tumor location, and maximum diameter to predict tumor classification and develop a nomogram. Receiver operator characteristic (ROC) curves were used to evaluate the ability of the nomogram to distinguish between two tumors with different risk classifications, and a calibration curve was used to evaluate the consistency between the predicted risk and the actual risk. The texture signature achieved high efficacy in predicting the mitotic index area under the curve ([AUC], 0.906; 95% confidence interval [CI]: 0.813, 0.961). A nomogram for prediction of the risk classification of GIST, which incorporated this texture signature together with maximum tumor diameter and location, allowed good discrimination in the training cohort (AUC, 0.878; 95% CI: 0.769, 0.960) and the validation cohort (AUC, 0.903; 95% CI: 0.732, 0.922). The texture-based model can be used to predict GIST mitotic index and risk classification preoperatively. 2. 3.

Clinical features, MRI, and 18F-FDG-PET in differential diagnosis of Parkinson disease from multiple system atrophy.

ObjectiveThis study aimed to differentiate the variations in the clinical characteristics, MRI irregularity, and glucose metabolism on 18F‐FDG‐PET for the differential diagnosis of Parkinson's Disease (PD), MSA with predominant Parkinsonism (MSA‐P), and MSA with predominant cerebellar features (MSA‐C).MethodsThirty PD patients, 22 MSA‐P patients, and 28 MSA‐C patients received an MRI and 20 PD patients, 11 MSA‐P patients, and 13 MSA‐C patients received 18F‐FDG‐PET.ResultsFirstly, we found that the clinical data presented a tremor at rest, bradykinesia, and postural instability that was predominated in PD (100%), MSA‐P (86.4%), and MSA‐C (53.6%) patients, respectively. Then, we used MRI analyses and found that putamina atrophy and hyperintensive rim (T2WI) were characteristic features in MSA‐P and cerebellar atrophy, the “hot cross bun” sign and signal rise in the middle cerebellar peduncle were more obvious in MSA‐C. To further explore the distinctions among the 3 diseases, we also used 18F‐FDG‐PET technology for our examination and found a decrease in glucose metabolism in the parietal area for Parkinson's Disease (PD), in the bilateral putamen for MSA‐P, and in the bilateral cerebellum for MSA‐C.ConclusionThis study identified the distinctive features of the clinic symptoms, MRI irregularity, and glucose metabolism on 18F‐FDG‐PET, which provided a new basis for the differential diagnosis of Parkinson's Disease (PD), MSA with predominant Parkinsonism (MSA‐P), and MSA with predominant cerebellar features (MSA‐C).

MRI in Children With Pyriform Sinus Fistula

A variety of imaging modalities have been described for the diagnosis of congenital pyriform sinus fistula (CPSF). To date, there have been few MRI reports. To evaluate MRI findings of CPSF and interobserver reliability. Retrospective. In all, 115 patients aged 23 days to 15.4 years at operation. 3.0T/axial T1 -weighted image (T1 WI)-SPIR, axial T2 WI, axial T2 WI-STIR, coronal T2 WI-SPIR, diffusion-weighted imaging (DWI), axial and coronal gadolinium-enhanced T1 WI-SPIR. For each patient, the medical records, including demographics, clinical manifestations, and MRI findings were reviewed. All the MRI studies were interpreted by three radiologists independently. Kendall's W test was made to determine the interobserver reliability of three reviewers for MRI findings. CPSF occurred on the left side in 104 (90.4%) patients and on the right side in 11 (9.6%) patients. The male-to-female ratio was 59:56. The age at first episode varied from birth to 12.3 years. There was one neonate patient, who presented with a unilocular cystic mass in the left neck. A tunnel-like lesion between the pyriform fossa and the upper pole of the thyroid gland, T2 high signal behind the cricothyroid joint, thyroid gland involvement, deep neck abscesses or masses were noted in 46 (40%), 93 (80.9%), 96 (83.5%), and 36 (31.3%) patients, respectively. There was excellent interobserver reliability for all the MRI findings, ranging from 0.84 to 1.00. The sinus tract presenting with a tunnel-like lesion goes behind the cricothyroid joint in most cases. For patients presenting with acute suppurative thyroiditis (AST) or neck infection with thyroid gland involvement, the presence of T2 high signal behind the cricothyroid joint highly suggests the diagnosis of CPSF. MRI is a reliable method for the diagnosis of CPSF. 4 TECHNICAL EFFICACY STAGE: 2.

Evaluation of renal ischemia-reperfusion injury by magnetic resonance imaging texture analysis: An experimental study.

To explore the value of MRI texture analysis in evaluating the presence and severity of early renal ischemia-reperfusion injury (IRI). Healthy New Zealand rabbits were used (IRI group, N=54; control group, N=8). Rabbits in the IRI group underwent left renal artery clamping for 60minutes. Magnetic resonance imaging was performed before and at 1, 12, 24, and 48hours after IRI. The relationship between MRI texture features and histopathology parameters was assessed using Pearson's correlation coefficients. The diagnostic performance of texture features in kidney differentiation at different time points was assessed by receiver operating characteristic curve analysis. T2 WI_S(3,-3)Inverse_Difference_Moment had the strongest correlation with brush border destruction, tubular epithelial edema, necrosis, and cast (r=0.56, -0.58, 0.62, and 0.69, respectively; all P<.001). BOLD_S(4,-4)Correlation had the strongest correlation with interstitial inflammatory cell infiltration (r=0.63, P<.001). SWI_S(4,4)Difference_Entropy had the strongest correlation with microvessel density (r=0.61, P<.001). The areas under the curve for T2 WI_S(3,-3)Inverse_Difference_Moment, SWI_S(4,4)Difference_Entropy, and BOLD_S(4,-4)Correlation in kidney differentiation before IRI and that at 1 and 12hours after reperfusion were 0.76, 0.72, and 0.70, respectively; the values before IRI and at 24 and 48hours after reperfusion were 0.84, 0.81, and 0.69, respectively. The area under the curve for T2 WI_S(3,-3)Inverse_Difference_Moment in kidney differentiation at 1 and 12hours after reperfusion and that at 24 and 48hours after reperfusion was 0.66. Magnetic resonance imaging texture analysis can be used for evaluating the presence and severity of early renal IRI.

Imaging Features and Floating-Ice Sign of Primary Lymphoma of the Bone

Background: The imaging presentation of primary bone lymphoma is unclear. Objectives: The present study aimed to investigate the imaging presentations of primary bone lymphoma especially a specific “floating-ice” sign. Patients and Methods: Forty one patients with primary bone lymphoma confirmed by pathology with 27 males and 14 females and an age range of 2 - 76 (mean = 40) years were enrolled. The clinical and imaging data were analyzed. Results: The tumor involved long bones in 17 cases, flat bones in 12, spine in eight and irregular bones in four cases. The imaging presentations were divided into five types: infiltrative type in nine cases (22%), osteolytic in 14 cases (34.1%), osteosclerotic in four cases (9.8%), mixed in 11 cases (26.8%) including four cases with a “floating-ice” sign and cystic in three cases (7.3%). In plain radiography, only three of four long bone lesions in children had a varying degree of periosteal reaction. Among 20 cases with CT scanning, sixteen had soft tissue masses, seventeen had ill-defined margins, and three had well-defined margins with sclerotic rims. Among twelve patients with MRI, ten had soft tissue masses with well-defined margins. MRI demonstrated a greater extent of lesion than CT. In MRI T1 weighted image (T1 WI), isointense signal was seen in three cases, hypointense signal in five and mixed signal in four. In T2 WI, isointense and hypointense signal was detected in five cases, hyperintense signal in three and mixed hyperintense signal in four. Conclusion: Primary bone lymphoma occurs most frequently in long and flat bones as infiltrative osteolytic destruction, and combined plain radiographs, CT and MRI help obtain a correct diagnosis.

Comparison of Biparametric and Multiparametric MRI for Clinically Significant Prostate Cancer Detection With PI-RADS Version 2.1.

Biparametric MRI (bpMRI) without dynamic contrast-enhanced MRI (DCE-MRI) results in an elimination of adverse events, shortened examination time, and reduced costs, compared to multiparametric MRI (mpMRI). The ability of bpMRI to detect clinically significant prostate cancer (csPC) with the Prostate Imaging and Reporting Data System version 2.1 (PI-RADS v2.1) compared to standard mpMRI has not been studied extensively. To compare the interobserver reliability and diagnostic performance for detecting csPC of bpMRI and mpMRI using PI-RADS v2.1. Retrospective. In all, 103 patients with elevated prostate-specific antigen (PSA) levels who underwent mpMRI and subsequent MRI-ultrasonography fusion-guided prostate-targeted biopsy (MRGB) with or without prostatectomy. T2 -weighted imaging (T2 WI), diffusion-weighted imaging (DWI), and DCE-MRI at 3T. Three readers independently assessed each suspected PC lesion, assigning a score of 1-5 for T2 WI, a score of 1-5 for DWI, and positive and negative for DCE-MRI according to PI-RADS v2.1 and determined the overall PI-RADS assessment category of bpMRI (T2 WI and DWI) and mpMRI (T2 WI, DWI, and DCE-MRI). The reference standard was MRGB or prostatectomy-derived histopathology. Statistical analysis was performed using the kappa statistic and McNemar and Delong tests. Of the 165 suspected PC lesions in 103 patients, 81 were diagnosed with csPC and 84 with benign conditions. Interobserver variability of PI-RADS assessment category showed good agreement for bpMRI (kappa value = 0.642) and mpMRI (kappa value = 0.644). For three readers, the diagnostic sensitivity was significantly higher for mpMRI than for bpMRI (P < 0.001 to P = 0.016, respectively), whereas diagnostic specificity was significantly higher for bpMRI than for mpMRI (P < 0.001 each). For three readers, the area under the receiver operating characteristic curve (AUC) was higher for bpMRI than for mpMRI; however, the difference was significant only for Reader 1 and Reader 3 (Reader 1: 0.823 vs. 0.785, P = 0.035; Reader 2: 0.852 vs. 0.829, P = 0.099; and Reader 3: 0.828 vs. 0.773, P = 0.002). For detecting csPC using PI-RADS v2.1, the interobserver reliability and diagnostic performance of bpMRI was comparable with those of mpMRI. 4 TECHNICAL EFFICACY STAGE: 2.

Automatic T Staging Using Weakly Supervised Deep Learning for Nasopharyngeal Carcinoma on MR Images.

Recent studies have shown that deep learning can help tumor staging automatically. However, automatic nasopharyngeal carcinoma (NPC) staging is difficult due to the lack of large and slice-level annotated datasets. To develop a weakly-supervised deep-learning method to predict NPC patients' T stage without additional annotations. Retrospective. In all, 1138 cases with NPC from 2010 to 2012 were enrolled, including a training set (n=712) and a validation set (n=426). 1.5T, T1 -weighted images (T1 WI), T2 -weighted images (T2 WI), contrast-enhanced T1 -weighted images (CE-T1 WI). We used a weakly-supervised deep-learning network to achieve automated T staging of NPC. T usually refers to the size and extent of the main tumor. The training set was employed to construct the deep-learning model. The performance of the automated T staging model was evaluated in the validation set. The accuracy of the model was assessed by the receiver operating characteristic (ROC) curve. To further assess the performance of the deep-learning-based T score, the progression-free survival (PFS) and overall survival (OS) were performed. The Sklearn package in Python was applied to calculate the area under the curve (AUC) of the ROC. The survcomp package was used for calculations and comparisons between C-indexes. The software SPSS was employed to conduct survival analysis and chi-square tests. The accuracy of the deep-learning model was 75.59% in the validation set. The average AUC of the ROC curve of different stages was 0.943. There were no significant differences in the C-indexes of PFS and OS from the deep-learning model and those from TNM staging, with P values of 0.301 and 0.425, respectively. This weakly-supervised deep-learning approach can perform fully automated T staging of NPC and achieve good prognostic performance. 3 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020;52:1074-1082.

Multiparametric MRI Radiomic Model for Preoperative Predicting WHO/ISUP Nuclear Grade of Clear Cell Renal Cell Carcinoma.

Nuclear grade is of importance for treatment selection and prognosis in patients with clear cell renal cell carcinoma (ccRCC). To develop and validate an MRI-based radiomic model for preoperative predicting WHO/ISUP nuclear grade in ccRCC. Retrospective. In all, 379 patients with histologically confirmed ccRCC. Training cohort (n = 252) and validation cohort (n = 127) were randomly assigned. Pretreatment 3.0T renal MRI. Imaging sequences were fat-suppressed T2 WI, contrast-enhanced T1 WI, and diffusion weighted imaging. Three prediction models were developed using selected radiomic features, radiomic and clinicoradiologic characteristics, and a model containing only clinicoradiologic characteristics. Receiver operating characteristic (ROC) curves and area under the curve (AUC) were used to assess the predictive performance of these models in predicting high-grade ccRCC. The least absolute shrinkage and selection operator (LASSO) and minimum redundancy maximum relevance (mRMR) method were used for the selection of radiomic features and clinicoradiologic characteristics, respectively. Multivariable logistic regression analysis was used to develop the radiomic signature of radiomic features and clinicoradiologic model of clinicoradiologic characteristics. The radiomic signature showed good performance in discriminating high-grade (grades 3 and 4) from low-grade (grades 1 and 2) ccRCC, with sensitivity, specificity, and AUC of 77.3%, 80.0%, and 0.842, respectively, in the validation cohort. The radiomic model, combining radiomic signature and clinicoradiologic characteristics, displayed good predictive ability for high-grade with sensitivity, specificity, and accuracy of 63.6%, 93.3%, and 88.2%, respectively, in the validation cohort. The radiomic model showed a significantly better performance than the clinicoradiologic model (P < 0.05). Multiparametric MRI-based radiomic model can predict WHO/ISUP grade in patients with ccRCC with satisfying performance, and thus could help the physician to improve treatment decisions. 3 TECHNICAL EFFICACY STAGE: 2.

Comparative Study of Amide Proton Transfer-Weighted Imaging and Intravoxel Incoherent Motion Imaging in Breast Cancer Diagnosis and Evaluation.

Amide proton transfer-weighted imaging (APTWI) and intravoxel incoherent motion imaging (IVIM) are valuable MRI techniques applied to cancer. To compare APTWI and IVIM in the diagnosis of benign and malignant breast lesions and to evaluate the correlations between different parameters (MTRasym [3.5 ppm], D, D*, and f) and prognostic factors for breast cancer. Retrospective. In all, 123 breast lesions were studied before treatment, including 58 benign lesions and 65 malignant lesions. Conventional MRI (T1 WI, T2 WI, and diffusion-weighted imaging [DWI]), APTWI, and IVIM MRI at 3T. The magnetization transfer ratio asymmetry at 3.5 ppm (MTRasym [3.5 ppm]), diffusion coefficient (D), pseudo diffusion coefficient (D*), and perfusion fraction (f) values were compared between the benign and malignant groups and between groups with different expression levels of prognostic factors. Individual sample t-test, χ2 test, Spearman correlation, logistic regression, and the Delong test. The D and MTRasym (3.5 ppm) values of the malignant group were lower than those of the benign group; however, D* and f values were higher than those of the benign group (all P < 0.05). The areas under the curve (AUCs) of D, MTRasym (3.5 ppm), D*, and f were 0.809, 0.778, 0.670, and 0.766, respectively; however, only the difference between AUC (D) and AUC (D*) was significant (Z = 2.374, P < 0.05). The D value showed a low correlation with the pathological grade and Ki-67 expression (| r | = 0.294, 0.367); the f value showed a low correlation with estrogen receptor (ER) expression (| r | = 0.382); and the MTRasym (3.5 ppm) value showed a low correlation with pathological grade (| r | = 0.371). This analysis revealed that both IVIM and APTWI could be used for the differential diagnosis of benign and malignant breast lesions, and APTWI-derived MTRasym (3.5 ppm), IVIM-derived D, D*, and f values showed correlations with some prognostic factors for breast cancer. 2 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:1175-1186.

Multiparametric MRI-Based Radiomics Nomogram for Predicting Lymphovascular Space Invasion in Endometrial Carcinoma.

Lymphovascular space invasion (LVSI) of endometrial carcinoma (EMC) is one of the important prognostic factors, which is not usually visible subjectively. Therefore, clinicians lack imaging-based evidence about LVSI for preoperative treatment decision-making. To develop a multiparametric MRI (mpMRI)-based radiomics nomogram for predicting LVSI in EMC and provide decision-making support to clinicians. Retrospective. In all, 144 patients with histologically confirmed EMC, 101 patients in a training cohort, and 43 patients in a test cohort. T2 WI, contrast enhanced-T1 WI, and diffusion-weighted imaging (DWI) at 3.0T MRI. Tumors were independently segmented images by two radiologists. Two pathologists reviewed the tissue specimens of the tumors to identify the existence of LVSI in consensus. The intraclass correlation coefficient was used to test the reliability and least absolute shrinkage and selection operator (LASSO) regression for features selection and then developed a radiomics signature named Rad-score. A nomogram was developed in the training cohort. The diagnostic performance of the nomogram model was assessed by area under the curve (AUC) of the receiver operator characteristic (ROC) in the training and test cohort, respectively. LVSI was identified in 32 patients (22.2%). Older age and high grade were correlated with LVSI at univariate analysis. There were five radiomics features that were identified as independent risk factors for LVSI by LASSO regression. Based on age, grade, and Rad-score, the AUC values of the nomogram model to predict LVSI in the training and test cohort were 0.820 (95% confidence interval [CI]: 0.725, 0.916; sensitivity: 82.6%, specificity: 72.9%), 0.807 (95% CI: 0.673, 0.941; sensitivity: 77.8%, specificity: 78.6%), respectively. The radiomic-based machine-learning model using a nomogram algorithm achieved high diagnostic performance for predicting LVSI of EMC preoperatively, which could enhance risk stratification and provide support for therapeutic decision-making. 2. 3. J. Magn. Reson. Imaging 2020;52:1257-1262.

Comparison of the Imaging and Pathological Features of Telangiectatic Osteosarcoma and Aneurysmal Bone Cyst of Long Tubular Bones

Background: The clinical, imaging and pathological features of telangiectatic osteosarcoma (TOS) in long bones are helpful for differentiation from aneurysmal bone cysts (ABC). Objectives: To investigate the clinical, imaging and pathological features of TOS in comparison with ABC in long bones. Patients and Methods: Seventeen patients with TOS and 17 patients with ABC in our hospital were analyzed in the clinical, imaging and pathological features for proper diagnosis. Results: For plain radiographs in 17 TOS cases, a soft tissue mass was detected in 14 cases, lytic bone destruction in 13, ill-defined border in 12, periosteal reaction in eight, formation of the Codman triangle in five, pathological fracture in three, and calcification in one. In 17 cases with ABC, an ill-defined border was found in fifteen cases, lytic and expansile destruction in eleven, thinned sclerotic rim in twelve, eccentrically location in ten, coarse septations in eight, a soft tissue mass in three, pathological fracture in three, periosteal reaction in two, and ill-defined border in two cases. In CT imaging of TOS, multiple cysts were in twelve cases, and fluid-fluid levels in three. In ABC, multiple cysts were in eleven cases while fluid-fluid levels in four. In MRI presentations of TOS, low and medium signal was present in the T1-weighted imaging (T1WI), whereas mixed (9/12) or high signal (3/12) was presented in the T2 WI with inhomogeneous signal. Four cases had multiple cysts and fluid-fluid levels in the bone destruction and soft tissues. The cysts had long T1 and long T2 signal with isointense signal in the septations (5/12). Fat inhibition sequence showed clear cystic changes. In ABC, low complete or incomplete signal was demonstrated in all lesions with a higher incidence of fluid-fluid levels (8/11). The lesion was of low and medium inhomogeneous signal in the T1 WI but mixed signal in T2 WI (6/11) including hyperintense signal in the upper part but mixed signal in the lower part in four cases or hyperintense signal (1/11) in T2 WI. The histopathological features of TOS were bleeding inside the lesion, highly atypia tumor cells of malignancy in the septa and tumorous bone tissue. The septa of ABC were composed of proliferative fibroblasts with dispersed polykaryocytes, reactive woven bone but no atypical cells. Conclusion: The clinical, imaging and pathological characteristics should be combined to make a correct differential diagnosis between telangiectatic osteosarcoma and aneurysmal bone cyst.

Predicting the Outcome of Transcatheter Arterial Embolization Therapy for Unresectable Hepatocellular Carcinoma Based on Radiomics of Preoperative Multiparameter MRI.

In unresectable hepatocellular carcinoma (HCC), methods to predict patients at increased risk of progression are required. To investigate the feasibility of radiomics model in predicting early progression of unresectable HCC after transcatheter arterial chemoembolization (TACE) therapy using preoperative multiparametric magnetic resonance imaging (MP-MRI). Retrospective. A total of 84 patients with BCLC B stage HCC from one medical center. According to the modified response evaluation criteria in solid tumors, patients who progressed at 6 months after TACE therapy were assigned as the progressive disease (PD) group (n = 32). Patients whose MRI was performed on four devices were divided into a training cohort (n = 67). Patients whose MRI was performed on other than the previous four devices were used as the testing set (n = 17). 3.0T, 1.5T axial T2 -weighted imaging (T2 WI), diffusion-weighted imaging (DWI, b = 0, 500 s/mm2 ), and apparent diffusion coefficient (ADC) ASSESSMENT: PD was confirmed via imaging studies with MRI. Risk factors, including age, alpha fetoprotein (AFP), size, and radiomic-related features of PD were assessed. In addition, the discrimination ability of each radiomics signature was tested on an independent testing set. The area under the receiver-operator characteristic (ROC) curve (AUC) was used to evaluate the predictive accuracy of the radiomic signature in both the training and testing sets. The results indicated that the MP-MRI model achieved the greatest benefit. In the testing set, the model based on DWI features presented an AUC of (b = 0, 0.786; b = 500, 0.729), followed by T2 WI features (0.729) and ADC (0.714). The AUC of the MP-MRI signature was increased to 0.800 compared to any single MRI signature. The multivariate logistic analysis identified the radiomics signature as independent parameters of PD, while clinical information such as age, AFP, size, etc., had no significance in the PD group. Preoperative MP-MRI has the potential to predict the outcome of TACE therapy for unresectable HCC. In addition, these image features may be complementary to the current staging systems of HCC patients. 2. 3. J. Magn. Reson. Imaging 2020;52:1083-1090.