Background: Accurate thoracolumbar fracture classification is central to treatment planning, but manual interpretation of CT images can be time-consuming and variable.

Objective: To develop a CT-based deep learning workflow for automated vertebral localization, fracture screening, AOSpine thoracolumbar ABC classification, and Osteoporotic Fracture (OF) grading.

Methods: This retrospective study included 845 spinal CT examinations with expert consensus labels. Total Segmentator was used for vertebral segmentation and level identification. Three-dimensional vertebral CT patches were then processed by cascaded 3D ResNet-18 models. The AOSpine model fused CT image features with four automatically extracted bone-void features: total void volume, void-volume ratio, void count, and maximum void volume. The OF model used a hierarchical multi-head structure for OF1–OF5 grading. Performance was assessed using Dice, level-identification accuracy, accuracy, precision, recall, specificity, F1-score, AUC, average precision, and quadratic weighted kappa.

Results: Total Segmentator achieved a mean Dice coefficient of 0.846 and vertebral level-identification accuracy of 92.22%. The fracture screening model achieved an accuracy of 0.988. For AOSpine ABC classification, the CT plus bone-void model achieved an overall accuracy of 62.5% and macro-F1 of 0.512. For OF grading, the hierarchical CT-only model achieved an accuracy of 0.737, macro-F1 of 0.692, quadratic weighted kappa of 0.639, macro-AUC of 0.894, and macro-AP of 0.767.

Conclusion: The proposed CT-based workflow demonstrated feasibility for automated vertebral segmentation, fracture screening, AOSpine ABC classification, and OF grading. Further multicenter validation is required before clinical deployment.

Purpose/Objective(s): The UICC and AJCC have recently added imaging-detected extranodal extension (iENE) as a cN modifier for the upcoming 9th edition TNM Classification (TNM9) of HPV-positive oropharyngeal carcinoma (HPV+ OPC). While academic radiologists demonstrate good inter-rater reliability in identifying imaging-detected extranodal extension (iENE), its reliability among community radiologists has not been studied. Therefore, we conducted an international study to assess the reliability and impact of training on iENE recognition by two groups of community radiologists in the USA and Canada.

Materials/Methods: Community radiologists from The Permanente Medical Group (TPMG) and a Quebec Radiology Group (QR) who responded to “Expression-of-Interest” emails were recruited. They were asked to consult training material addressing the Head-and-Neck Cancer-International Group consensus definitions of iENE status before proceeding with a Round-1 (20 cases) and Round-2 (30 cases) iENE review. After each round, expert interpretations were provided to both groups for self-reflection. The TPMG group also had an online group review following Round-1. Gwet’s AC1 concordance score was estimated for the overall, TPMG, and QR groups.

Results: A total of 10 radiologists (5 each from TPMG and QR groups) were recruited. The mean (standard deviation) agreement for Round-1 and Round-2 was 86.0% (7.4) and 86.0% (6.0), respectively. The Gwet’s AC1 concordance score were 0.72 (0.50-0.93) (moderate) and 0.76 (0.62-0.89) (substantial) in Round-1 and Round-2, respectively. Gwet’s AC1 score in Round-1 vs Round-2 was 0.74 (0.53-0.95) (moderate) vs 0.82 (0.69-0.94) (substantial) for the TPMG group, and 0.72 (0.48-0.97) (moderate) vs 0.69 (0.50-0.88)] (moderate) for the QR group, respectively.

Conclusion: This study shows good inter-rater reliability for iENE status among community radiologists after applying the Head and-Neck-Cancer-International Group consensus definitions, and using high diagnostic certainty. With adherence to guidelines, wider dissemination of HPV+ OPC prognostic models that include iENE status appear feasible. Educational materials are also available to further augment reproducibility.

Chronic constipation is widespread. It is believed that between 10% and 15% of the population are treated for constipation on an ongoing basis.

Before the introduction of modern treatments, acute ischemic stroke (AIS) resulted in 10% early mortality, around 50% of survivors left with moderate-to-severe neurologic deficits, and 25% left dependent on others. This drastically improved with the introduction of intravenous tissue plasminogen activators and later with endovascular treatment (EVT). Patient selection for EVT relies on dedicated multimodality neuroimaging conducted with four main goals – 1) exclude a hemorrhagic stroke and identify early ischemic changes, 2) identify a proximal large vessel occlusion, 3) determine the volume of ‘ischemic core’, and 4) determine the volume of ‘ischemic penumbra’. This comparative narrative review aims to discuss in detail how different imaging modalities are used in the context of AIS to select patients for EVT. This includes computed tomography (CT) and magnetic resonance imaging (MRI), including their role in angiographic and perfusion imaging. Based on the success of EVT trials from 2015 and 2018, the updated American Heart Association - American Stroke Association guidelines state that non-contrast head CT and CT angiography are sufficient to identify patients who are fit to undergo EVT in the early window (

Interventional Radiology (IR) as a speciality presents the need for a focused specialist training pathway