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.

Aims: To compare the diagnostic performance of microvascular Doppler ultrasonography (MVUS) to B-mode and con-ventional colour Doppler US (CDUS) for detecting acute pyelonephritis (APN) lesions in children. Material and methods: An IRB-approved retrospective study was performed. From July 2018 to January 2019, 41 APN lesions in 28 children (15 boys, 13 girls; age range, 1-196 months; mean age, 53 months) who underwent 99mTc‒dimercaptosuccinic acid renal scintig-raphy (DMSA) or contrast-enhanced computed tomography (CECT) and US including B-mode, CDUS, and MVUS were enrolled in this study. Three paediatric radiologists independently reviewed the B-mode, CDUS and MVUS images for the DMSA or CECT-proven APN lesions and evaluated the lesion visibility, lesion distinguishability and diagnostic confidence between the MVUS and CDUS images. Results: A total 41 of APN lesions were verified by DMSA (41 lesions) or CECT (3 lesions) during the same hospitalization period with renal US. Among 41 APN lesions, 52.8% was visible on B-mode, 85.4% on CDUS, and 94.3% on MVUS (p<0.001). Comparing the extent and margins of the lesions, MVUS had better results than CDUS in 41.5% of the lesions, CDUS had better results in 6.5% and they were equal in 52% (p<0.001). The diagnostic con-fidence of the APN lesions was higher for MVUS than CDUS in 36.6%, higher for CDUS than MVUS in 4.9%, and equal in the remaining 58.5% (p<0.05). The interobserver agreement was fair to moderate. Conclusions: MVUS showed improved detectability of hypoperfused areas in paediatric APN and provided higher diagnostic confidence.

Gastrointestinal (GI) emergencies in neonates and infants encompass from the beginning to the end of the GI tract. Both congenital and acquired conditions can cause various GI emergencies in neonates and infants. Given the overlapping or nonspecific clinical findings of many different neonatal and infantile GI emergencies and the unique characteristics of this age group, appropriate imaging is key to accurate and timely diagnosis while avoiding unnecessary radiation hazard and medical costs. In this paper, we discuss the radiological findings of essential neonatal and infantile GI emergencies, including esophageal atresia and tracheoesophageal fistula, hypertrophic pyloric stenosis, duodenal atresia, malrotation, midgut volvulus for upper GI emergencies, and jejunoileal atresia, meconium ileus, meconium plug syndrome, meconium peritonitis, Hirschsprung disease, anorectal malformation, necrotizing enterocolitis, and intussusception for lower GI emergencies.

목적 코로나바이러스감염증–19 영향으로 소아청소년과 내원 환자 수가 현저히 감소한 2020년에 소아 장중첩증에는 어떤 변화가 발생했는지 알아보고자 한다. 대상과 방법 2011년부터 2020년까지 고려대학교 안산병원에 내원하여 장중첩증 진단 후 관장정복술을 받은 18세 미만의 환자의 성별, 나이, 관장정복술의 연도별/월별/계절별 빈도, 관장정복술의 연도별/월별 실패율, 장중첩증의 연도별/월별 재발률을 조사하였다. 또한 2011–2019년에 내원한 환자군과 2020년에 내원한 환자군 사이에 평균 나이, 관장정복술 실패율 및 장중첩증 재발률에서 통계적 차이가 있는지 분석하였다. 결과 10년간 859건의 관장정복술은 남아에서 더 많이 시행되었고 평균 나이는 22.2개월, 연령별로는 1세 미만이 가장 많았다. 연도별 빈도는 2014년에 가장 높았고 2020년에 가장 낮았으며, 월별 빈도는 12월과 9월이 높았고 3월이 가장 낮았다. 2020년 환자군( n = 27)은 2011–2019년 환자군( n = 832)에 비해 평균 나이가 어리고(18.1개월 vs. 22.8개월) 관장정복술 실패율이 높고 (7.4% vs. 2.4%) 장중첩증 재발률이 높았으나(14.8% vs. 7.3%) 통계적 유의성을 보이지는 않았다( p = 0.07, p = 0.15, p = 0.14). 결론 코로나바이러스감염증–19가 유행한 2020년에는 장중첩증으로 관장정복술을 받은 환자 수가 급격히 감소하였고, 지난 10년의 기록과 비교할 때 환자의 나이가 더 어린 경향을 보였으며, 관장정복술의 실패율과 장중첩증 재발률은 가장 높은 수치를 기록하였다.