Purpose: To determine whether machine learning assisted-texture analysis of multi-energy virtual monochromatic image (VMI) datasets from dual-energy CT (DECT) can be used to differentiate metastatic head and neck squamous cell carcinoma (HNSCC) lymph nodes from lymphoma, inflammatory, or normal lymph nodes. Materials and methods: A retrospective evaluation of 412 cervical nodes from 5 different patient groups (50 patients in total) having undergone DECT of the neck between 2013 and 2015 was performed: (1) HNSCC with pathology proven metastatic adenopathy, (2) HNSCC with pathology proven benign nodes (controls for (1)), (3) lymphoma, (4) inflammatory, and (5) normal nodes (controls for (3) and (4)). Texture analysis was performed with TexRAD® software using two independent sets of contours to assess the impact of inter-rater variation. Two machine learning algorithms (Random Forests (RF) and Gradient Boosting Machine (GBM)) were used with independent training and testing sets and determination of accuracy, sensitivity, specificity, PPV, NPV, and AUC. Results: In the independent testing (prediction) sets, the accuracy for distinguishing different groups of pathologic nodes or normal nodes ranged between 80 and 95%. The models generated using texture data extracted from the independent contour sets had substantial to almost perfect agreement. The accuracy, sensitivity, specificity, PPV, and NPV for correctly classifying a lymph node as malignant (i.e. metastatic HNSCC or lymphoma) versus benign were 92%, 91%, 93%, 95%, 87%, respectively. Conclusion: Machine learning assisted-DECT texture analysis can help distinguish different nodal pathology and normal nodes with a high accuracy.

Accidental breach of the vertebral artery (VA) during the performance of cervical pain blocks can result in significant morbidity. Whereas anatomical variations have been described for the foraminal (V2) segment of the VA, those involving its V3 portion (between the C2 transverse process and dura) have not been investigated and may be of importance for procedures targeting the third occipital nerve or the lateral atlantoaxial joint. Five hundred computed tomography angiograms of the neck performed in patients older than 50 years for the management of cerebrovascular accident or cervical trauma (between January 2010 and May 2016) were retrospectively and independently reviewed by 2 neuroradiologists. Courses of the VA in relation to the lateral aspect of the C2/C3 joint and the posterior surface of the C1/C2 joint were examined. For the latter, any medial encroachment of the VA (or one of its branches) was noted. The presence of a VA loop between C1 and C2 and its distance from the upper border of the superior articular process (SAP) of C3 were also recorded. If the VA loop coursed posteriorly, its position in relation to 6 fields found on the lateral aspects of the articular pillars of C2 and C3 was tabulated. At the C1/C2 level, the VA coursed medially over the lateral quarter of the dorsal joint surface in 1% of subjects (0.6% and 0.4% on the left and right sides, respectively; P = 0.998). A VA loop originating between C1 and C2 was found to travel posteroinferiorly over the anterolateral aspect of the inferior articular pillar of C2 in 55.5% of patients on the left and 41.9% on the right side (P < 0.001), as well as over the SAP of C3 in 0.4% of subjects. When present in the quadrant immediately cephalad to the C3 SAP, VA loops coursed within 2.0 ± 1.5 and 3.3 ± 2.5 mm on the left and right sides, respectively, of its superior aspect (P < 0.001). The VA commonly travels adjacent to areas targeted by third occipital nerve procedures and more rarely over the access point for lateral atlantoaxial joint injections. Modifications to existing techniques may reduce the risk of accidental VA breach.

Current radiomic studies of head and neck squamous cell carcinomas (HNSCC) are typically based on datasets combining tumors from different locations, assuming that the radiomic features are similar based on histopathologic characteristics. However, molecular pathogenesis and treatment in HNSCC substantially vary across different tumor sites. It is not known if a statistical difference exists between radiomic features from different tumor sites and how they affect machine learning model performance in endpoint prediction. To answer these questions, we extracted radiomic features from contrast-enhanced neck computed tomography scans (CTs) of 605 patients with HNSCC originating from the oral cavity, oropharynx, and hypopharynx/larynx. The difference in radiomic features of tumors from these sites was assessed using statistical analyses and Random Forest classifiers on the radiomic features with 10-fold cross-validation to predict tumor sites, nodal metastasis, and HPV status. We found statistically significant differences (p-value ≤ 0.05) between the radiomic features of HNSCC depending on tumor location. We also observed that differences in quantitative features among HNSCC from different locations impact the performance of machine learning models. This suggests that radiomic features may reveal biologic heterogeneity complementary to current gold standard histopathologic evaluation. We recommend considering tumor site in radiomic studies of HNSCC.