Deep Learning-Assisted Diagnosis of Pediatric Skull Fractures on Plain Radiographs.

OBJECTIVE: To develop and evaluate a deep learning-based artificial intelligence (AI) model for detecting skull fractures on plain radiographs in children. MATERIALS AND METHODS: This retrospective multi-center study consisted of a development dataset acquired from two hospitals (n = 149 and 264) and an external test set (n = 95) from a third hospital. Datasets included children with head trauma who underwent both skull radiography and cranial computed tomography (CT). The development dataset was split into training, tuning, and internal test sets in a ratio of 7:1:2. The reference standard for skull fracture was cranial CT. Two radiology residents, a pediatric radiologist, and two emergency physicians participated in a two-session observer study on an external test set with and without AI assistance. We obtained the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity along with their 95% confidence intervals (CIs). RESULTS: The AI model showed an AUROC of 0.922 (95% CI, 0.842-0.969) in the internal test set and 0.870 (95% CI, 0.785-0.930) in the external test set. The model had a sensitivity of 81.1% (95% CI, 64.8%-92.0%) and specificity of 91.3% (95% CI, 79.2%-97.6%) for the internal test set and 78.9% (95% CI, 54.4%-93.9%) and 88.2% (95% CI, 78.7%-94.4%), respectively, for the external test set. With the model's assistance, significant AUROC improvement was observed in radiology residents (pooled results) and emergency physicians (pooled results) with the difference from reading without AI assistance of 0.094 (95% CI, 0.020-0.168; p = 0.012) and 0.069 (95% CI, 0.002-0.136; p = 0.043), respectively, but not in the pediatric radiologist with the difference of 0.008 (95% CI, -0.074-0.090; p = 0.850). CONCLUSION: A deep learning-based AI model improved the performance of inexperienced radiologists and emergency physicians in diagnosing pediatric skull fractures on plain radiographs.

A preliminary study of shear-wave elastography for the evaluation of varicocele in adolescents and young adults.

PURPOSE: The purpose of this study was to use shear-wave elastography (SWE) to assess testicular stiffness changes during the Valsalva maneuver in adolescents and young adults with varicocele, to compare these changes according to clinical severity, and to evaluate the role of SWE in the diagnosis of varicocele. METHODS: This study included patients undergoing testicular ultrasonography for the diagnosis of varicocele or for post-varicocelectomy follow-up between June 2016 and February 2017. Fifty-four testicles of 27 consecutive patients (mean age, 15.9 years) were classified by clinical grade (grade 0-3). Using SWE, mean testicular stiffness (Emean) was measured at rest and during the Valsalva maneuver. The correlations between multiple ultrasonographic parameters (volume asymmetry, Emean at rest, and absolute and percentage changes in Emean during the Valsalva maneuver) and clinical grade were assessed using the Spearman correlation test. RESULTS: The Emean at rest was similar across clinical grades. During the Valsalva maneuver, increased testicular stiffness was frequently observed in patients with grade 2 or 3 varicocele but rarely observed in those with grade 0 or 1 varicocele. The changes in Emean were positively correlated with the clinical grade (all P<0.001), whereas the Emean at rest and volume asymmetry were not (all P>0.05). CONCLUSION: A transient, reversible increase in testicular stiffness during the Valsalva maneuver was observed in adolescents with high-grade varicocele, and the degree of stiffness change was correlated with the clinical grade. Stiffness change identified using SWE during the Valsalva maneuver is a potential surrogate indicator of venous congestion.

Shear-wave elastography for the assessment of testicular involvement of hematologic malignancies in children and young adults: a feasibility study.

PURPOSE: This study aimed to assess the feasibility of shear-wave elastography (SWE) for assessing the testicular involvement of hematologic malignancies in children and young adults. METHODS: Eight patients (mean age, 11.0 years; range, 0.8 to 20 years) with biopsy-confirmed testicular involvement of hematologic malignancy between January 2018 and December 2020 were retrospectively evaluated. Multiparametric ultrasound examinations, including grayscale, color Doppler ultrasonography (CDUS), and SWE, were performed. Stiffness was measured in the involved testicular area and contralateral normal parenchyma. If there was bilateral testicular involvement, the stiffness of the involved area and the adjacent normal echoic parenchyma was measured on one testis. The Mann-Whitney U test was used to compare stiffness values. RESULTS: On grayscale, the testicular lesions were noted as a solitary mass in one patient, multiple lesions in four patients, and diffuse involvement in three patients. On CDUS and SWE, all patients demonstrated increased vascularity, and the stiffness of the involved area was higher than the values of normal parenchyma (the involved area vs. normal parenchyma, 11.6 kPa [3.9-20.2 kPa] vs. 2.9 kPa [1.1-3.7 kPa], P=0.003). The ratio of stiffness between the involved area and normal parenchyma was 3.4, ranging from 1.9 to 5.1. One patient showed decreased stiffness on follow-up SWE after treatment (affected testis vs. normal testis: initial, 13.8 vs. 3.2 kPa; 1 year later, 2.2 vs. 2.4 kPa). CONCLUSION: Increased testicular stiffness on SWE in children and young adults with hematologic malignancies suggests the possibility of testicular involvement.

Transcatheter Dynamic Contrast-Enhanced MR Lymphangiography for Nontraumatic Lymphatic Disorders: Technical Feasibility and Imaging Findings.

This study aimed to evaluate the feasibility and added value of transcatheter dynamic contrast-enhanced magnetic resonance (MR) lymphangiography for nontraumatic lymphatic disorders. Five patients (2 males and 3 females; median age, 16.0 years; range, 3-74 years) who underwent both intranodal and transcatheter dynamic contrast-enhanced MR lymphangiography for suspected nontraumatic lymphatic leakages from June 2017 to January 2020 were included in this retrospective study. The imaging findings of both dynamic contrast-enhanced MR lymphangiography techniques were assessed for the presence of chylolymphatic reflux or direct sign of leakage. Intranodal dynamic contrast-enhanced MR lymphangiography demonstrated chylolymphatic reflux into the thoracic area in 2 patients (40%) but no direct evidence of leakage in any of the 5 patients. Transcatheter dynamic contrast-enhanced MR lymphangiography revealed chylolymphatic reflux and extravasation of the contrast agent in all 5 patients (100%). In conclusion, transcatheter dynamic contrast-enhanced MR lymphangiography may reveal additional signs of reflux and extravasation even when the findings of intranodal dynamic contrast-enhanced MR lymphangiography are negative.

Generating synthetic contrast enhancement from non-contrast chest computed tomography using a generative adversarial network.

This study aimed to evaluate a deep learning model for generating synthetic contrast-enhanced CT (sCECT) from non-contrast chest CT (NCCT). A deep learning model was applied to generate sCECT from NCCT. We collected three separate data sets, the development set (n = 25) for model training and tuning, test set 1 (n = 25) for technical evaluation, and test set 2 (n = 12) for clinical utility evaluation. In test set 1, image similarity metrics were calculated. In test set 2, the lesion contrast-to-noise ratio of the mediastinal lymph nodes was measured, and an observer study was conducted to compare lesion conspicuity. Comparisons were performed using the paired t-test or Wilcoxon signed-rank test. In test set 1, sCECT showed a lower mean absolute error (41.72 vs 48.74; P < .001), higher peak signal-to-noise ratio (17.44 vs 15.97; P < .001), higher multiscale structural similarity index measurement (0.84 vs 0.81; P < .001), and lower learned perceptual image patch similarity metric (0.14 vs 0.15; P < .001) than NCCT. In test set 2, the contrast-to-noise ratio of the mediastinal lymph nodes was higher in the sCECT group than in the NCCT group (6.15 ± 5.18 vs 0.74 ± 0.69; P < .001). The observer study showed for all reviewers higher lesion conspicuity in NCCT with sCECT than in NCCT alone (P ≤ .001). Synthetic CECT generated from NCCT improves the depiction of mediastinal lymph nodes.

Posterior Lung Herniation in Pulmonary Agenesis and Aplasia: Chest Radiograph and Cross-Sectional Imaging Correlation.

OBJECTIVE: To describe the anatomic locations and imaging features of posterior lung herniation in unilateral pulmonary agenesis and aplasia, focusing on radiograph-CT/MRI correlation. MATERIALS AND METHODS: A total of 10 patients (seven with pulmonary agenesis and three with pulmonary aplasia, male: female = 1:9, mean age 7.3 years, age range from 1 month to 20 years) were included. Chest radiographs (n = 9), CT (n = 9), and MRI (n = 1) were reviewed to assess the type of lung underdevelopment, presence of anterior and posterior lung herniation, bronchus origin, supplying artery, and draining vein of the herniated lung. RESULTS: Pulmonary agenesis/aplasia more commonly affected the left lung (n = 7) than the right lung (n = 3). Anterior lung herniation was observed in nine of the 10 patients. Posterior lung herniation was observed in seven patients with left pulmonary agenesis/aplasia. Two patients showed posterior lung herniation crossing the midline but not beyond the aorta, and five patients showed the posteriorly herniated right lower lobe crossing the midline to extend into the left hemithorax farther beyond the descending thoracic aorta through the space between the esophagus and the aorta. This anatomical configuration resulted in a characteristic radiographic finding of a radiolucent area with a convex lateral border and a vertical medial border in the left lower lung zone, revealing a tongue-like projection on CT and MRI. CONCLUSION: Posterior lung herniation occurs in unilateral left lung agenesis/aplasia. Approximately 70% of the cases of posterior lung herniation reveal a unique radiolucent tongue-like projection in the left lower lung zone on imaging studies, which is caused by the extension of the posteriorly herniated right lung farther beyond the descending aorta.

Monitoring Cerebral Perfusion Changes Using Arterial Spin-Labeling Perfusion MRI after Indirect Revascularization in Children with Moyamoya Disease.

OBJECTIVE: To assess the role of arterial spin-labeling (ASL) perfusion MRI in identifying cerebral perfusion changes after indirect revascularization in children with moyamoya disease. MATERIALS AND METHODS: We included pre- and postoperative perfusion MRI data of 30 children with moyamoya disease (13 boys and 17 girls; mean age ± standard deviation, 6.3 ± 3.0 years) who underwent indirect revascularization between June 2016 and August 2017. Relative cerebral blood flow (rCBF) and qualitative perfusion scores for arterial transit time (ATT) effects were evaluated in the middle cerebral artery (MCA) territory on ASL perfusion MRI. The rCBF and relative time-to-peak (rTTP) values were also measured using dynamic susceptibility contrast (DSC) perfusion MRI. Each perfusion change on ASL and DSC perfusion MRI was analyzed using the paired t test. We analyzed the correlation between perfusion changes on ASL and DSC images using Spearman's correlation coefficient. RESULTS: The ASL rCBF values improved at both the ganglionic and supraganglionic levels of the MCA territory after surgery (p = 0.040 and p = 0.003, respectively). The ATT perfusion scores also improved at both levels (p < 0.001 and p < 0.001, respectively). The rCBF and rTTP values on DSC MRI showed significant improvement at both levels of the MCA territory of the operated side (all p < 0.05). There was no significant correlation between the improvements in rCBF values on the two perfusion images (r = 0.195, p = 0.303); however, there was a correlation between the change in perfusion scores on ASL and rTTP on DSC MRI (r = 0.701, p < 0.001). CONCLUSION: Recognizing the effects of ATT on ASL perfusion MRI may help monitor cerebral perfusion changes and complement quantitative rCBF assessment using ASL perfusion MRI in patients with moyamoya disease after indirect revascularization.

The usefulness of noninvasive liver stiffness assessment using shear-wave elastography for predicting liver fibrosis in children.

BACKGROUND: Pediatric patients with liver disease require noninvasive monitoring to evaluate the risk of fibrosis progression. This study aimed to identify the significant factors affecting liver stiffness values using two-dimensional shear-wave elastography (2D-SWE), and determine whether liver stiffness can predict the fibrosis stage of various childhood liver diseases. METHODS: This study included 30 children (22 boys and 8 girls; mean age, 5.1 ± 6.1 years; range, 7 days-17.9 years) who had undergone biochemical evaluation, 2D-SWE examination, histopathologic analysis of fibrosis grade (F0 to F3), assessment of necroinflammatory activity, and steatosis grading between August 2016 and March 2020. The liver stiffness from 2D-SWE was compared between fibrosis stages using Kruskal-Wallis analysis. Factors that significantly affected liver stiffness were evaluated using univariate and multivariate linear regression analyses. The diagnostic performance was determined from the area under the receiver operating curve (AUC) values of 2D-SWE liver stiffness. RESULTS: Liver stiffness at the F0-1, F2, and F3 stages were 7.9, 13.2, and 21.7 kPa, respectively (P < 0.001). Both fibrosis stage and necroinflammatory grade were significantly associated with liver stiffness (P < 0.001 and P = 0.021, respectively). However, in patients with alanine aminotransferase (ALT) levels below 200 IU/L, the only factor affecting liver stiffness was fibrosis stage (P = 0.030). The liver stiffness value could distinguish significant fibrosis (≥ F2) with an AUC of 0.950 (cutoff value, 11.3 kPa) and severe fibrosis (F3 stage) with an AUC of 0.924 (cutoff value, 18.1 kPa). The 2D-SWE values for differentiating significant fibrosis were 10.5 kPa (≥ F2) and 18.1 kPa (F3) in patients with ALT levels below 200 IU/L. CONCLUSION: The liver stiffness values on 2D-SWE can be affected by both fibrosis and necroinflammatory grade and can provide excellent diagnostic performance in evaluating the fibrosis stage in various pediatric liver diseases. However, clinicians should be mindful of potential confounders, such as necroinflammatory activity or transaminase level, when performing 2D-SWE measurements for liver fibrosis staging.

Korean Society of Thyroid Radiology Guidelines for the Management of Pediatric Thyroid Nodules: Suitability and Risk Factors.

Background: The guideline for managing pediatric thyroid nodules has not been established. We investigated the appropriateness of the Korean Society of Thyroid Radiology (KSThR) guidelines and the need to consider risk factors for managing pediatric thyroid nodules. Methods: From August 2007 to February 2020, a database of thyroid nodules in patients younger than 19 years who underwent fine-needle aspiration (FNA) was analyzed. Two radiologists retrospectively reviewed ultrasound images to characterize the nodules according to the Korean Thyroid Imaging Reporting and Data System (K-TIRADS) criteria. Thyroid nodules were divided into benign and malignant nodules. Clinical risk factors were identified, including familial thyroid cancer, history of radiation therapy, and underlying thyroiditis. According to the K-TIRADS categories and the presence of risk factors, malignancy rates were calculated. We compared the diagnostic performance of the original KSThR guidelines and newly suggested criteria for FNA derived by simulation tests with variable size cutoffs in the total study population, a group with risk factors, and a group without risk factors. Results: A total of 107 patients (mean age, 13.9 years; range, 4-18 years; 83 females; 50 patients with risk factors) with 133 nodules (71 benign and 62 malignant) were included. The malignancy rate of thyroid nodules was higher in the group with risk factors (64.0%) than in the group without risk factors (43.9%; p = 0.037). Compared with the KSThR guideline (≥1.0 cm for K-TIRADS 4 and 5), the accuracy was higher (62.4% vs. 56.4%) and the unnecessary biopsy rate was lower (44.0% vs. 47.4%) when the new criteria (≥0.6 cm for K-TIRADS 4 and ≥0.5 cm for K-TIRADS 5) were used in the total study population. In particular, in the group with risk factors, the newly suggested size criteria had a much higher accuracy (73.1% vs. 59.7%) and lower unnecessary biopsy rate (30.4% vs. 35.5%) than the KSThR guideline. Conclusions: For the management of thyroid nodules in children, it is necessary to consider risk factors and adjust the criteria values for FNA. FNA should be considered in children with risk factors for thyroid cancer even if thyroid nodules are small.

MRI risk factors for development of avascular necrosis after closed reduction of developmental dysplasia of the hip: Predictive value of contrast-enhanced MRI.

PURPOSE: The purpose of this study was to identify imaging risk factors on contrast-enhanced hip MRI after closed reduction of developmental dysplasia of the hip (DDH) that could predict future development of avascular necrosis (AVN) of the femoral head. MATERIALS AND METHODS: Fifty-eight infants (F: M = 53: 5, aged 3-18 months) who underwent immediate postoperative contrast-enhanced hip MRI after closed reduction of DDH were included in this study. Quality of reduction (concentric vs eccentric reduction with or without obstacles), abduction angle of the hip, presence of ossific nucleus, and pattern of contrast enhancement of the femoral head were retrospectively evaluated on MRI. Interobserver agreement of contrast enhancement pattern on MRI were evaluated by two radiologists. Development of AVN was determined through radiographic findings at 1 year after reduction. RESULTS: AVN of the femoral head developed in 13 (22%) of 58 patients. Excessive abduction of the hip joint (OR 4.65, [95% CI 1.20, 18.06] and global decreased enhancement of the femoral head (OR 71.66, [95% CI 10.54, 487.31]) exhibited statistically significant differences between the AVN and non-AVN groups (P < 0.05). Eccentric reduction (P = 0.320) did not show statistically significant difference between two groups and invisible ossific nucleus (P = 0.05) showed borderline significance. Multi-variable logistic regression indicated that global decreased enhancement of the femoral head was a significant risk factor of AVN (OR 27.92, 95% CI [4.17, 350.18]) (P = 0.0031). Interobserver agreement of contrast enhancement pattern analysis and diagnosis of AVN were good (0.66, 95% CI [0.52, 0.80]). CONCLUSION: Contrast-enhanced hip MRI provides accurate anatomical assessment of the hip after closed reduction of DDH. Global decreased enhancement of the femoral head could be used as a good predictor for future development of AVN after closed reduction of DDH.

Evaluation of frequency-selective non-linear blending technique on brain CT in postoperative children with Moyamoya disease.

OBJECTIVE: To evaluate whether a frequency-selective non-linear blending (BC) technique can improve tissue contrast and infarct detection on non-enhanced brain CT (NECT) in postoperative Moyamoya (MMD) patients. MATERIALS AND METHODS: From January 2010 to December 2017, 33 children (13boys and 20girls; mean age 9.1±3.4 years) with MMD postoperatively underwent NECT followed by diffusion MRI. We compared the contrast-to-noise ratio (CNR) between gray matter (GM) and white matter (WM) in NECT and BC images and the CNR between the infarct lesion and adjacent normal-appearing brain in NECT and BC images using a paired t-test. We assessed image noise, GM-WM differentiation, artifacts, and overall quality using a Wilcoxon signed rank test. A McNemar two-tailed test was conducted to compare the diagnostic accuracy of infarct detection. RESULTS: The CNR between GM and WM and the CNR of the infarct was better in BC images than in NECT images (3.9±1.0 vs. 1.8±0.6, P<0.001 and 3.6±0.3 vs. 1.9±0.2, P<0.001), with no difference in overall image quality observed. The sensitivity and specificity of infarct detection were 55.0% and 76.9% using NECT, and 70.0% and 69.2% using BC technique. The diagnostic accuracy of NECT and BC technique was 63.6% (21/33) and 69.7% (23/33), respectively. CONCLUSION: This study showed that the BC technique improved CNR and maintained image quality. This technique may also be used to identify ischemic brain changes in postoperative MMD patients by improving the CNR of the infarct lesion.

Diagnosis of Hippocampal Sclerosis in Children: Comparison of Automated Brain MRI Volumetry and Readers of Varying Experience.

BACKGROUND. Hippocampal sclerosis (HS) is a leading cause of medically refractory temporal lobe epilepsy in children. The diagnosis is clinically important because most patients with HS have good postsurgical outcomes. OBJECTIVE. This study aimed to compare the performance of a fully automated brain MRI volumetric tool and readers of varying experience in the diagnosis of pediatric HS. METHODS. This retrospective study included 22 children with HS diagnosed between January 2009 and January 2020 who underwent surgery and an age- and sex-matched control group of 44 patients with normal MRI findings and extratemporal epilepsy diagnosed between January 2009 and January 2020. Regional brain MRI volumes were calculated from a high-resolution 3D T1-weighted sequence using an automated volumetric tool. Four readers (two pediatric radiologists [experienced] and two radiology residents [inexperienced]) visually assessed each MRI examination to score the likelihood of HS. One inexperienced reader repeated the evaluations using the volumetric tool. The area under the ROC curve (AUROC), sensitivity, and specificity for HS were computed for the volumetric tool and the readers. Diagnostic performances were compared using McNemar tests. RESULTS. In the HS group, the hippocampal volume (affected vs unaffected, 3.54 vs 4.59 cm3) and temporal lobe volume (affected vs unaffected, 5.66 vs 6.89 cm3) on the affected side were significantly lower than on the unaffected side (p < .001) using the volu-metric tool. AUROCs of the volumetric tool were 0.813-0.842 in patients with left HS and 0.857-0.980 in patients with right HS (sensitivity, 81.8-90.9%; specificity, 70.5-95.5%). No significant difference (p = .63 to > .99) was observed between the performance of the volumetric tool and the performance of the two experienced readers as well as one inexperienced reader (AUROCs for these three readers, 0.968-0.999; sensitivity, 86.4-90.9%; specificity, 100.0%). The volumetric tool had better performance (p < .001) than the other inexperienced reader (AUROC, 0.806; sensitivity, 81.8%; specificity, 47.7%). With subsequent use of the tool, this inexperienced reader showed a nonsignificant increase (p = .10) in AUROC (0.912) as well as in sensitivity (86.4%) and specificity (84.1%). CONCLUSION. A fully automated volumetric brain MRI tool outperformed one of two inexperienced readers and performed as well as two experienced readers in identifying and lateralizing HS in pediatric patients. The tool improved the performance of an inexperienced reader. CLINICAL IMPACT. A fully automated volumetric tool facilitates diagnosis of HS in pediatric patients, especially for an inexperienced reader.

Noise reduction approach in pediatric abdominal CT combining deep learning and dual-energy technique.

OBJECTIVES: To evaluate the image quality of low iodine concentration, dual-energy CT (DECT) combined with a deep learning-based noise reduction technique for pediatric abdominal CT, compared with standard iodine concentration single-energy polychromatic CT (SECT). METHODS: From December 2016 to May 2017, DECT with 300 mg•I/mL contrast medium was performed in 29 pediatric patients (17 boys, 12 girls; age, 2-19 years). The DECT images were reconstructed using a noise-optimized virtual monoenergetic reconstruction image (VMI) with and without a deep learning method. SECT images with 350 mg•I/mL contrast medium, performed within the last 3 months before the DECT, served as reference images. The quantitative and qualitative parameters were compared using paired t tests and Wilcoxon signed-rank tests, and the differences in radiation dose and total iodine administration were assessed. RESULTS: The linearly blended DECT showed lower attenuation and higher noise than SECT. The 60-keV VMI showed an increase in attenuation and higher noise than SECT. The combined 60-keV VMI plus deep learning images showed low noise, no difference in contrast-to-noise ratios, and overall image quality or diagnostic image quality, but showed a higher signal-to-noise ratio in the liver and lower enhancement of lesions than SECT. The overall image and diagnostic quality of lesions were maintained on the combined noise reduction approach. The CT dose index volume and total iodine administration in DECT were respectively 19.6% and 14.3% lower than those in SECT. CONCLUSION: Low iodine concentration DECT, combined with deep learning in pediatric abdominal CT, can maintain image quality while reducing the radiation dose and iodine load, compared with standard SECT. KEY POINTS: • An image noise reduction approach combining deep learning and noise-optimized virtual monoenergetic image reconstruction can maintain image quality while reducing radiation dose and iodine load. • The 60-keV virtual monoenergetic image reconstruction plus deep learning images showed low noise, no difference in contrast-to-noise ratio, and overall image quality, but showed a higher signal-to-noise ratio in the liver and a lower enhancement of lesion than single-energy polychromatic CT. • This combination could offer a 19.6% reduction in radiation dose and a 14.3% reduction in iodine load, in comparison with a control group that underwent single-energy polychromatic CT with the standard protocol.

Diagnostic Performance of a New Convolutional Neural Network Algorithm for Detecting Developmental Dysplasia of the Hip on Anteroposterior Radiographs.

OBJECTIVE: To evaluate the diagnostic performance of a deep learning algorithm for the automated detection of developmental dysplasia of the hip (DDH) on anteroposterior (AP) radiographs. MATERIALS AND METHODS: Of 2601 hip AP radiographs, 5076 cropped unilateral hip joint images were used to construct a dataset that was further divided into training (80%), validation (10%), or test sets (10%). Three radiologists were asked to label the hip images as normal or DDH. To investigate the diagnostic performance of the deep learning algorithm, we calculated the receiver operating characteristics (ROC), precision-recall curve (PRC) plots, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) and compared them with the performance of radiologists with different levels of experience. RESULTS: The area under the ROC plot generated by the deep learning algorithm and radiologists was 0.988 and 0.988-0.919, respectively. The area under the PRC plot generated by the deep learning algorithm and radiologists was 0.973 and 0.618-0.958, respectively. The sensitivity, specificity, PPV, and NPV of the proposed deep learning algorithm were 98.0, 98.1, 84.5, and 99.8%, respectively. There was no significant difference in the diagnosis of DDH by the algorithm and the radiologist with experience in pediatric radiology (p = 0.180). However, the proposed model showed higher sensitivity, specificity, and PPV, compared to the radiologist without experience in pediatric radiology (p < 0.001). CONCLUSION: The proposed deep learning algorithm provided an accurate diagnosis of DDH on hip radiographs, which was comparable to the diagnosis by an experienced radiologist.

Validation and feasibility of liver T1 mapping using free breathing MOLLI sequence in children and young adults.

We investigated the feasibility of free-breathing modified Look-Locker inversion recovery (MOLLI) sequence for measuring hepatic T1 values in children and young adults. To investigate the accuracy and the reproducibility of the T1 maps, a phantom study was performed with 12 different gadoterate meglumine concentrations and the T1 relaxation times of phantoms measured with the MOLLI sequence were compared against those measured with three different sequences: spin-echo inversion recovery, variable flip angle (VFA), and VFA with B1 correction. To evaluate the feasibility of free-breathing MOLLI sequence, hepatic T1 relaxation times obtained by free-breathing and breath-hold technique in twenty patients were compared. The phantom study revealed the excellent accuracy and reproducibility of MOLLI. In twenty patients, the mean value of hepatic T1 values obtained by free-breathing (606.7 ± 64.5 ms) and breath-hold (609.8 ± 64.0 ms) techniques showed no significant difference (p > 0.05). The Bland-Altman plot between the free-breathing and breath-hold revealed that the mean difference of T1 values was - 3.0 ms (- 0.5%). Therefore, T1 relaxation times obtained by MOLLI were comparable to the values obtained using the standard inversion recovery method. The hepatic T1 relaxation times measured by MOLLI technique with free-breathing were comparable to those obtained with breath-hold in children and young adults.

Clinicopathological findings of pediatric NTRK fusion mesenchymal tumors.

BACKGROUND: While ETV6- NTRK3 fusion is common in infantile fibrosarcoma, NTRK1/3 fusion in pediatric tumors is scarce and, consequently, not well known. Herein, we evaluated for the presence of NTRK1/3 fusion in pediatric mesenchymal tumors, clinicopathologically and immunophenotypically. METHODS: We reviewed nine NTRK fusion-positive pediatric sarcomas confirmed by fluorescence in situ hybridization and/or next-generation sequencing from Seoul National University Hospital between 2002 and 2020. RESULTS: One case of TPR-NTRK1 fusion-positive intracranial, extra-axial, high-grade undifferentiated sarcoma (12-year-old boy), one case of LMNA-NTRK1 fusion-positive low-grade infantile fibrosarcoma of the forehead (3-year-old boy), one case of ETV6-NTRK3 fusion-positive inflammatory myofibroblastic tumor (IMT) (3-months-old girl), and six cases of ETV6-NTRK3 fusion-positive infantile fibrosarcoma (median age: 2.6 months, range: 1.6-5.6 months, M: F = 5:1) were reviewed. The Trk immunopositivity patterns were distinct, depending on what fusion genes were present. We observed nuclear positivity in TPR-NTRK1 fusion-positive sarcoma, nuclear membrane positivity in LMNA-NTRK1 fusion-positive sarcoma, and both cytoplasmic and nuclear positivity in ETV6-NTRK3 fusion-positive IMT and infantile fibrosarcomas. Also, the TPR-NTRK1 fusion-positive sarcoma showed robust positivity for CD34/nestin, and also showed high mitotic rate. The LMNA-NTRK1 fusion-positive sarcoma revealed CD34/S100 protein/nestin/CD10 coexpression, and a low mitotic rate. The IMT with ETV6-NTRK3 fusion expressed SMA. Six infantile fibrosarcomas with ETV6-NTRK3 fusion showed variable coexpression of nestin (6/6)/CD10 (4/5)/ S100 protein (3/6). CONCLUSIONS: All cases of NTRK1 and NTRK3 fusion-positive pediatric tumors robustly expressed the Trk protein. A Trk immunopositive pattern and CD34/S100/nestin/CD10/SMA immunohistochemical expression may suggest the presence of NTRK fusion partner genes. LMNA-NTRK1 fusion sarcoma might be a low-grade subtype of infantile fibrosarcoma. Interestingly, more than half of the infantile fibrosarcoma cases were positive for S100 protein and CD10. The follow-up period of TPR-NTRK1 and LMNA-NTRK1 fusion-positive tumors are not enough to predict prognosis. However, ETV6-NTRK3 fusion-positive infantile fibrosarcomas showed an excellent prognosis with no evidence of disease for an average of 11.7 years, after gross total resection of the tumor.

Incidence and Risk Factors of Nausea and Vomiting after Exposure to Low-Osmolality Iodinated Contrast Media in Children: A Focus on Preparative Fasting.

OBJECTIVE: To evaluate the incidence and risk factors of emetic complications associated with the intravenous administration of low-osmolality iodinated contrast media (ICM) in children undergoing computed tomography (CT). MATERIALS AND METHODS: All children who underwent contrast-enhanced CT between April 2017 and July 2019 were included. Pediatric patients were instructed on the preparative dietary protocol at our institution. Experienced nurses in the radiology department monitored the children during the CT scans and recorded any emetic complications in their electronic medical records. These data were used to calculate the incidence of emetic complications. Various patient factors and technical factors, including fasting duration, the type and volume of ICM, and ongoing chemotherapy, were evaluated to identify risk factors for emetic complications using univariate and multivariate logistic regression analyses. RESULTS: Among the 864 children (mean age, 8.4 ± 5.7 years) evaluated, 18 (2.1%) experienced emetic complications (6 experienced nausea only and 12 experienced nausea and vomiting). None of the children developed aspiration pneumonia. The mean fasting duration of patients with emesis was 7.9 ± 5.7 hours (range, 3-21 hours), whereas that of patients without nausea was 8.7 ± 5.7 hours (range, 0-24 hours). Fasting duration was not associated with the development of nausea and vomiting (p = 0.634). Multivariate logistic regression analysis revealed that ongoing chemotherapy (odds ratio [OR] = 4.323; 95% confidence interval [CI] = 1.430-13.064; p = 0.009), iomeprol use (OR = 7.219; 95% CI = 1.442-36.146; p = 0.016), and iohexol use (OR = 5.241; 95% CI = 1.350-20.346; p = 0.017) were independent risk factors for emetic complications. CONCLUSION: Only a small proportion (2.1%) of children experienced nausea or vomiting after exposure to low-osmolality ICM. Many children underwent excessive fasting; however, fasting duration was not associated with nausea and vomiting. Moreover, ongoing chemotherapy and the use of iomeprol or iohexol were identified as potential risk factors for emetic complications in children.

Quantitative evaluation of Crohn's disease using dynamic contrast-enhanced MRI in children and young adults.

OBJECTIVES: To evaluate the clinical usefulness of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in children and young adults with Crohn's disease. METHODS: From August 2017 to October 2018, 30 patients with Crohn's disease (21 males and 9 females; mean age 15.1 ± 2.5 years) underwent DCE-MRI with MRI enterography. We assessed the endoscopic finding, pediatric Crohn's disease activity index (PCDAI), C-reactive protein (CRP) level (mg/dL), Crohn's disease MR index (CDMI) score, and the perfusion parameters of DCE-MRI (Ktrans, Kep, and Ve) at the ileocecal region between the inactive and active groups based on the histopathologic status. RESULTS: The active Crohn's disease group showed higher PCDAI, CRP, and CDMI scores than the inactive group (22.2 ± 18.8 vs. 6.3 ± 6.4, p = 0.027; 1.32 ± 1.79 vs. 0.10 ± 0.13, p = 0.005; 7.4 ± 3.9 vs. 4.5 ± 3.0, p = 0.047, respectively). The active Crohn's disease group showed a higher Ktrans value than the inactive group (0.31  ±  0.12 vs. 0.16 ±  0.46 min-1, p = 0.002). Endoscopic finding; PCDAI, CRP, and CDMI scores; and Ktrans value were significant parameters in the identification of the active Crohn's disease (p = 0.002, p < 0.001, p = 0.029, p = 0.006, and p < 0.001, respectively). Ktrans value was the most significant value for identifying active Crohn's disease in the multivariate logistic regression analysis (p = 0.013). CONCLUSION: Ktrans value could discriminate between inactive and active Crohn's diseases. Ktrans value may have the potential to monitor the pediatric Crohn's disease activity. KEY POINTS: • With dynamic contrast-enhanced MRI, we can quantitatively monitor the Crohn's disease status in pediatric patients and provide proper management plans to clinicians. • The Ktransvalue of dynamic contrast-enhanced MRI perfusion parameter, as well as the clinical pediatric Crohn's disease activity index, C-reactive protein level, the endoscopic score, and the Crohn's disease MR index, was higher in the active Crohn's disease than in the inactive group based on the histopathologic status. • The Ktransvalue among the dynamic contrast-enhanced MRI perfusion parameters was the most significant differentiating parameter for the active Crohn's disease from inactive status among those parameters (p = 0.013).

Application of T1-weighted BLADE sequence to abdominal magnetic resonance imaging of young children: a comparison with turbo spin echo sequence.

BACKGROUND: The image quality of abdominal magnetic resonance imaging (MRI) in children who cannot hold their breath has been severely impaired by motion artifacts. PURPOSE: To evaluate the usefulness of T1-weighted (T1W) BLADE MRI for axial abdominal imaging in children who cannot hold their breath. MATERIAL AND METHODS: Two different BLADE sequences, with and without an inversion recovery (IR-BLADE), were compared to conventional turbo-spin echo (TSE) with a high number of excitations in 18 consecutive patients who cannot hold their breath. Overall image quality, motion artifact, radial artifact, hepatic vessel sharpness, renal corticomedullary differentiation, and lesion conspicuity were retrospectively assessed by two radiologists, using 4- or 5-point scoring systems. Signal variations of each sequence were measured for a quantitative comparison. The acquisition times of the three sequences were compared. RESULTS: IR-BLADE and BLADE showed significantly improved overall image quality and reduced motion artifact compared with TSE. IR-BLADE showed significantly better hepatic vessel sharpness and corticomedullary differentiation compared to both BLADE and TSE. Radial artifacts were only observed on IR-BLADE and BLADE. In nine patients with lesions, there were no significant differences in lesion conspicuity among three sequences. Compared to TSE, both IR-BLADE and BLADE showed decreased signal variations in the liver and muscle, and an increased signal variation through air. The mean acquisition times for IR-BLADE, BLADE, and TSE were comparable. CONCLUSION: Compared to the TSE sequence, T1W IR-BLADE for pediatric abdominal MRI resulted in improved image quality, tissue contrast with a diminished respiratory motion artifact, and a comparable acquisition time.

Application of Vendor-Neutral Iterative Reconstruction Technique to Pediatric Abdominal Computed Tomography.

OBJECTIVE: To compare image qualities between vendor-neutral and vendor-specific hybrid iterative reconstruction (IR) techniques for abdominopelvic computed tomography (CT) in young patients. MATERIALS AND METHODS: In phantom study, we used an anthropomorphic pediatric phantom, age-equivalent to 5-year-old, and reconstructed CT data using traditional filtered back projection (FBP), vendor-specific and vendor-neutral IR techniques (ClariCT; ClariPI) in various radiation doses. Noise, low-contrast detectability and subjective spatial resolution were compared between FBP, vendor-specific (i.e., iDose1 to 5; Philips Healthcare), and vendor-neutral (i.e., ClariCT1 to 5) IR techniques in phantom. In 43 patients (median, 14 years; age range 1-19 years), noise, contrast-to-noise ratio (CNR), and qualitative image quality scores of abdominopelvic CT were compared between FBP, iDose level 4 (iDose4), and ClariCT level 2 (ClariCT2), which showed most similar image quality to clinically used vendor-specific IR images (i.e., iDose4) in phantom study. Noise, CNR, and qualitative imaging scores were compared using one-way repeated measure analysis of variance. RESULTS: In phantom study, ClariCT2 showed noise level similar to iDose4 (14.68-7.66 Hounsfield unit [HU] vs. 14.78-6.99 HU at CT dose index volume range of 0.8-3.8 mGy). Subjective low-contrast detectability and spatial resolution were similar between ClariCT2 and iDose4. In clinical study, ClariCT2 was equivalent to iDose4 for noise (14.26-17.33 vs. 16.01-18.90) and CNR (3.55-5.24 vs. 3.20-4.60) (p > 0.05). For qualitative imaging scores, the overall image quality ([reader 1, reader 2]; 2.74 vs. 2.07, 3.02 vs. 2.28) and noise (2.88 vs. 2.23, 2.93 vs. 2.33) of ClariCT2 were superior to those of FBP (p < 0.05), and not different from those of iDose4 (2.74 vs. 2.72, 3.02 vs. 2.98; 2.88 vs. 2.77, 2.93 vs. 2.86) (p > 0.05). CONCLUSION: Vendor-neutral IR technique shows image quality similar to that of clinically used vendor-specific hybrid IR technique for abdominopelvic CT in young patients.