Implications of Pediatric Artificial Intelligence Challenges for Artificial Intelligence Education and Curriculum Development.

Several radiology artificial intelligence (AI) courses are offered by a variety of institutions and educators. The major radiology societies have developed AI curricula focused on basic AI principles and practices. However, a specific AI curriculum focused on pediatric radiology is needed to offer targeted education material on AI model development and performance evaluation. There are inherent differences between pediatric and adult practice patterns, which may hinder the application of adult AI models in pediatric cohorts. Such differences include the different imaging modality utilization, imaging acquisition parameters, lower radiation doses, the rapid growth of children and changes in their body composition, and the presence of unique pathologies and diseases, which differ in prevalence from adults. Thus, to enhance radiologists' knowledge of the applications of AI models in pediatric patients, curricula should be structured keeping in mind the unique pediatric setting and its challenges, along with methods to overcome these challenges, and pediatric-specific data governance and ethical considerations. In this report, the authors highlight the salient aspects of pediatric radiology that are necessary for AI education in the pediatric setting, including the challenges for research investigation and clinical implementation.

Deep learning-based prediction of rib fracture presence in frontal radiographs of children under two years of age: a proof-of-concept study.

OBJECTIVE: In this proof-of-concept study, we aimed to develop deep-learning-based classifiers to identify rib fractures on frontal chest radiographs in children under 2 years of age. METHODS: This retrospective study included 1311 frontal chest radiographs (radiographs with rib fractures, n = 653) from 1231 unique patients (median age: 4 m). Patients with more than one radiograph were included only in the training set. A binary classification was performed to identify the presence or absence of rib fractures using transfer learning and Resnet-50 and DenseNet-121 architectures. The area under the receiver operating characteristic curve (AUC-ROC) was reported. Gradient-weighted class activation mapping was used to highlight the region most relevant to the deep learning models' predictions. RESULTS: On the validation set, the ResNet-50 and DenseNet-121 models obtained an AUC-ROC of 0.89 and 0.88, respectively. On the test set, the ResNet-50 model demonstrated an AUC-ROC of 0.84 with a sensitivity of 81% and specificity of 70%. The DenseNet-50 model obtained an AUC of 0.82 with 72% sensitivity and 79% specificity. CONCLUSION: In this proof-of-concept study, a deep learning-based approach enabled the automatic detection of rib fractures in chest radiographs of young children with performances comparable to pediatric radiologists. Further evaluation of this approach on large multi-institutional data sets is needed to assess the generalizability of our results. ADVANCES IN KNOWLEDGE: In this proof-of-concept study, a deep learning-based approach performed well in identifying chest radiographs with rib fractures. These findings provide further impetus to develop deep learning algorithms for identifying rib fractures in children, especially those with suspected physical abuse or non-accidental trauma.