The Value of Computed Tomography in Differentiating Heterotopic Pancreas from Small Gastrointestinal Stromal Tumor.

BACKGROUND: It is critical for an accurate preoperative diagnosis of heterotopic pancreas (HP) and small gastrointestinal stromal tumor (GIST), given the unique treatment and prognosis of the two tumors. This study aims to investigate HP's computed tomography (CT) features and identify the distinguishing characteristics between HP and small GIST. METHODS: From January 2016 to August 2020, our hospital database was searched for confirmed histopathological results and CT scans for HP and GIST for further analysis. The statistically significant variables were determined by using Fisher's exact test, the Mann-Whitney U test, the receiver operating characteristic (ROC) curve and the inverse probability weighting method. RESULTS: CT images and clinical data were reviewed for 24 participants with HP and 34 patients with small GIST. Contour, border, relative enhancement grade, surface dimple, duct-like structure, short diameter (SD), attenuation of each lesion in the unenhanced phase (Lp), and the enhancement ratio of tumor in the venous phase (ER) were significant for differentiating HP from small GIST. Threshold values for SD and Lp were 1.40 cm and 42.33 Hounsfield units, respectively. Ill-defined border, surface dimple, ductlike structure, and Lp were independent factors that differentiated HP from small GIST. Additionally, SD and ER were also found to be independent factors. CONCLUSIONS: Contour, relative enhancement grade, SD, and Lp could effectively differentiate HP from small GIST, demonstrating improved diagnostic performance compared to other parameters. The presence of ductlike structures and surface dimples could further characterize HP. These findings may help distinguish HP from small GIST and avoid unnecessary invasive examination and therapy in individuals with asymptomatic HP.

Clot burden of acute pulmonary thromboembolism: comparison of two deep learning algorithms, Qanadli score, and Mastora score.

BACKGROUND: The deep learning convolution neural network (DL-CNN) benefits evaluating clot burden of acute pulmonary thromboembolism (APE). Our objective was to compare the performance of the deep learning convolution neural network trained by the fine-tuning [DL-CNN (ft)] and the deep learning convolution neural network trained from the scratch [DL-CNN (fs)] in the quantitative assessment of APE. METHODS: We included the data of 680 cases for training DL-CNN by DL-CNN (ft) and DL-CNN (fs), then retrospectively included 410 patients (137 patients with APE, 203 males, mean age 60.3±11.4 years) for testing the models. The distribution and volume of clots were respectively assessed by DL-CNN(ft) and DL-CNN(fs), and sensitivity, specificity, and area under the curve (AUC) were used to evaluate their performances in detecting clots on a per-patient and clot level. Radiologists evaluated the distribution of clots, Qanadli score, and Mastora score and right ventricular metrics, and the correlation of clot volumes with right ventricular metrics were analyzed with Spearman correlation analysis. RESULTS: On a per-patient level, the two DL-CNN models had high sensitivities and moderate specificities [DL-CNN (ft): 100% and 77.29%; DL-CNN (fs): 100% and 75.82%], and their AUCs were comparable (Z=0.30, P=0.38). On a clot level, DL-CNN (ft) and DL-CNN (fs) sensitivities and specificities in detecting central clots were 99.06% and 72.61%, and 100% and 70.63%, respectively. DL-CNN (ft) sensitivities and specificities in detecting peripheral clots were mostly higher than those of DL-CNN (fs), and their AUCs were comparable. Clot volumes measured with the two models were similar (U=85094.500, P=0.741), and significantly correlated with Qanadli scores [DL-CNN(ft) r=0.825, P<0.001, DL-CNN(fs) r=0.827, P<0.001] and Mastora scores [DL-CNN(ft) r=0.859, P<0.001, DL-CNN(fs) r=0.864, P<0.001]. Clot volumes were also correlated with right ventricular metrics. Clot burdens were increased in the low-risk, moderate-risk, and high-risk patients. Binary logistic regression revealed that only the ratio of right ventricular area/left ventricular area (RVa/LVa) was an independent predictor of in-hospital death (odds ratio 6.73; 95% CI, 2.7-18.12, P<0.001). CONCLUSIONS: Both DL-CNN (ft) and DL-CNN (fs) have high sensitivities and moderate specificities in detecting clots associated with APE, and their performances are comparable. While clot burdens quantitatively calculated by the two DL-CNN models are correlated with right ventricular function and risk stratification, RVa/LVa is an independent prognostic factor of in-hospital death in patients with APE.

Utility of intravoxel incoherent motion MRI derived parameters for prediction of aggressiveness in urothelial bladder carcinoma.

BACKGROUND: Preoperative accurate judgment of aggressiveness is of great importance to determine treatment and prognosis of bladder cancers. PURPOSE: To evaluate the utility of IVIM-MRI parameters in predicting aggressiveness of bladder urothelial carcinoma. STUDY TYPE: Prospective. POPULATION MODEL: Sixty-seven patients with bladder urothelial cancer. FIELD STRENGTH/SEQUENCE: 3.0 T/T2WI and IVIM-MRI. ASSESSMENT: All cases were categorized in low-, intermediate-, or high-aggressiveness proposed by Kobayashi depending on the T stage and pathological grade. Images analysis and IVIM-derived parameters (apparent diffusion coefficient standard ADC, true diffusion coefficient D, pseudodiffusion coefficient D*, and perfusion fraction f) measurements were performed independently by two radiologists. STATISTICAL TESTS: Comparisons of IVIM-derived parameters in different aggressiveness levels were performed using one-way analysis of variance or Kruskal-Wallis test. Binary logistic regression models were used to calculate predicted probability of combined parameters. Diagnostic performance of individual and combined parameters for distinguishing high- from low-/intermediate-aggressiveness was assessed by using the receiver operating characteristics (ROC) curve. RESULTS: The ADC and D values differed significantly among low-, intermediate-, and high-aggressive urothelial bladder carcinoma, respectively (P < 0.05). The f value showed significant differences between low- and high-aggressive and between intermediate- and high-aggressive bladder carcinoma (P < 0.05). The best parameter for differentiating high- from low-/intermediate-aggressive urothelial bladder carcinoma was ADC value, with the area under ROC curve (AUC) and accuracy of 0.895 and 85.97%, followed by f and D values with AUCs of 0.873 and 0.862, respectively. The best combination of parameters was combined D and f values, with AUC and accuracy of 0.931 and 91.82%. DATA CONCLUSION: ADC value showed slightly better diagnostic performance than D and f values in predicting bladder cancer aggressiveness. The combination of D and f model can produce a robust value than single parameter in evaluating aggressiveness. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2018;48:1648-1656.

Renal epithelioid angiomyolipoma: magnetic resonance imaging characteristics.

OBJECTIVE: The aim of the study was to analyze MR imaging features of renal epithelioid angiomyolipoma (EAML). METHODS: This study included 17 patients with histopathologically confirmed renal EAML who underwent renal MRI scanning before radical or partial nephrectomy. MR images were retrospectively reviewed and correlated with pathological findings. RESULT: Fifteen lesions (88.2%) appeared as round or oval. The tumor-kidney interface was round in 14 lesions (82.4%). Fifteen tumors (88.2%) presented mainly isointensity on T1WI, and eleven tumors (64.7%) presented mainly hypointensity on T2WI. Twelve lesions (70.6%) showed restricted diffusion on DWI, and the mean ADC value was 1.23 ± 0.28 × 10-3mm2/s. Minimal fat component was identified as loss of signal intensity on opposed-phase MR images in 6 cases (35.3%). Sixteen lesions (100%) demonstrated inhomogeneous enhancement, and 7 of 16 masses (43.8%) showed reticular enhancement. Rapid wash-in and wash-out enhancement was seen in 13 masses (81.3%). In the corticomedullary phase, the mass showed markedly enhancement in 14 cases (87.5%). The irregular vessels and hemorrhage were detected in 4 cases (23.5%) and 7 cases (41.2%), respectively. One patient (5.9%) had a lymph node involvement at initial diagnosis, and showed distant metastasis after operation. In the immunohistochemical analysis, 15 tumors (88.2%) were positive for melanocytic marker (HMB45 or Melan-A), and all cases (100%) were negative for epithelial-associated markers (CK or AE1/AE3). CONCLUSION: The presence of hypointensity on T2WI, restricted diffusion on DWI, round tumor-kidney interface, reticular, and marked enhancement (rapid wash-in and wash-out) should further raise suspicion for renal EAML. The diagnosis may be confirmed by pathological analysis.