Purpose: This study aimed to determine whether zero echo time magnetic resonance imaging (ZTE-MRI), as an alternative imaging modality, and conventional computed tomography (CT) have similar diagnostic qualities for assessing pediatric osseous pathologies. Materials and Methods: Twenty-six sets of pediatric musculoskeletal CT and MRI scans (15 boys and 11 girls; mean age, 12 ± 4 years; range, 5–23 years) acquired at Seoul National University Children’s Hospital (January 2021 to November 2023) were retrospectively evaluated. CT-like images from ZTE-MRI were generated using grayscale inversion. Two radiologists independently assessed ZTE-MRI image quality (S anat) on a 5-point scale (1 = nondiagnostic, 5 = excellent) and a comparative scale (–2 = CT greater, 0 = same, 2 = ZTE-MRI greater) for lesion delineation (Scomp). The confidence interval of proportions and intraclass correlation coefficient were calculated to assess inter-rater agreement, and Wilcoxon rank-sum test, Mann–Whitney U test, or paired t-test was used to compare image quality or cortical thickness between the modalities. Results: ZTE-MRI demonstrated diagnostic quality (S anat ≥ 3) in 85%–96% of the cases, 89%–96% for cortical delineation, 92%–100% for intramedullary cavity (IMC) delineation, and 92% for lesion delineation. Compared with conventional CT, ZTE-MRI showed comparable diagnostic power (Scomp ≥ –1) in 92%–96% of the cases, with Scomp scores indicating no significant difference in lesion delineation (p = 0.53 in reader 1 and p = 0.25 in reader 2). There was a preference for CT over ZTE-MRI in terms of overall image quality and delineation of the cortex and IMC (p < 0.001). Cortical thickness was not significantly different (p = 0.11) between ZTE-MRI and CT. Conclusion ZTE-MRI demonstrated diagnostic quality comparable to that of CT, particularly in lesion delineation. In addition to the unique information that conventional MRI can provide, ZTE-MRI can provide additional information about osseous structures similar to that provided by CT, which we believe will be valuable in the future.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study aimed to determine whether zero echo time magnetic resonance imaging (ZTE-MRI), as an alternative imaging modality, and conventional computed tomography (CT) have similar diagnostic qualities for assessing pediatric osseous pathologies. Materials and Methods: Twenty-six sets of pediatric musculoskeletal CT and MRI scans (15 boys and 11 girls; mean age, 12 ± 4 years; range, 5–23 years) acquired at Seoul National University Children’s Hospital (January 2021 to November 2023) were retrospectively evaluated. CT-like images from ZTE-MRI were generated using grayscale inversion. Two radiologists independently assessed ZTE-MRI image quality (S anat) on a 5-point scale (1 = nondiagnostic, 5 = excellent) and a comparative scale (–2 = CT greater, 0 = same, 2 = ZTE-MRI greater) for lesion delineation (Scomp). The confidence interval of proportions and intraclass correlation coefficient were calculated to assess inter-rater agreement, and Wilcoxon rank-sum test, Mann–Whitney U test, or paired t-test was used to compare image quality or cortical thickness between the modalities. Results: ZTE-MRI demonstrated diagnostic quality (S anat ≥ 3) in 85%–96% of the cases, 89%–96% for cortical delineation, 92%–100% for intramedullary cavity (IMC) delineation, and 92% for lesion delineation. Compared with conventional CT, ZTE-MRI showed comparable diagnostic power (Scomp ≥ –1) in 92%–96% of the cases, with Scomp scores indicating no significant difference in lesion delineation (p = 0.53 in reader 1 and p = 0.25 in reader 2). There was a preference for CT over ZTE-MRI in terms of overall image quality and delineation of the cortex and IMC (p < 0.001). Cortical thickness was not significantly different (p = 0.11) between ZTE-MRI and CT. Conclusion ZTE-MRI demonstrated diagnostic quality comparable to that of CT, particularly in lesion delineation. In addition to the unique information that conventional MRI can provide, ZTE-MRI can provide additional information about osseous structures similar to that provided by CT, which we believe will be valuable in the future.

In the rapidly evolving healthcare environment, radiologists strive to establish their rightful place.Thus, there is a need for enhanced outpatient and clinical education within the Department of Radiology and exploration of its methodologies. Accordingly, the Korean Society of Radiology established a task force to investigate the clinical and outpatient practice status of radiologists overseas, current state of related education, involvement of other specialties in radiologic practices and education in Korea, and clinical and outpatient practice status among Korean radiologists. Furthermore, a survey on clinical competency enhancement was conducted among the members of the Korean Society of Radiology. These findings suggest the need for visibility and clinical competency enhancement in radiologists and methodologies for strengthening clinical competencies.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Purpose: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. Methods: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. Results: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. Conclusion The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.