BACKGROUND AND PURPOSE: Although intracranial atherosclerotic disease is often encountered during endovascular treatment for acute vertebrobasilar occlusions, its clinical implication is not well-known. We aimed to evaluate whether intracranial atherosclerotic disease influences the clinical outcomes following endovascular treatment of acute vertebrobasilar occlusive stroke. MATERIALS AND METHODS: Fifty-one patients with acute vertebrobasilar occlusive stroke were included. The onset-to-groin puncture time was ≤12 hours, and aspiration- or stent-based thrombectomy was used as the primary treatment method. Following primary endovascular treatment, intracranial atherosclerotic disease (IAD group) was angiographically diagnosed when a fixed focal stenosis was observed at the occlusion site, whereas embolism (embolic group) was diagnosed if no stenosis was observed. Clinical and treatment variables were compared in both groups, and IAD was evaluated as a prognostic factor for clinical outcomes. RESULTS: The baseline NIHSS score tended to be lower (14 versus 22, P = .097) in the IAD group (n = 19) than in the embolic group (n = 32). The procedural time was longer in the IAD group (96 versus 61 minutes, P = .002), despite similar rates of TICI 2b–3 (89.5% versus 87.5%, P = 1.000). The NIHSS score at 7 days was higher (21 versus 8, P = .060) and poor outcomes (mRS 4–6 at 3 months) were more frequent in the IAD group (73.7% versus 43.8%, P = .038). IAD (odds ratio, 5.469; 95% CI, 1.09–27.58; P = .040) was independently associated with poor outcomes. CONCLUSIONS: An arterial occlusion related to IAD was associated with a longer procedural time and poorer clinical outcome. Further studies are warranted to elucidate the appropriate endovascular strategy.

Acute intracranial haemorrhage (AIH) is a potentially life-threatening emergency that requires prompt and accurate assessment and management. This study aims to develop and validate an artificial intelligence (AI) algorithm for diagnosing AIH using brain-computed tomography (CT) images. A retrospective, multi-reader, pivotal, crossover, randomised study was performed to validate the performance of an AI algorithm was trained using 104,666 slices from 3010 patients. Brain CT images (12,663 slices from 296 patients) were evaluated by nine reviewers belonging to one of the three subgroups (non-radiologist physicians, n = 3; board-certified radiologists, n = 3; and neuroradiologists, n = 3) with and without the aid of our AI algorithm. Sensitivity, specificity, and accuracy were compared between AI-unassisted and AI-assisted interpretations using the chi-square test. Brain CT interpretation with AI assistance results in significantly higher diagnostic accuracy than that without AI assistance (0.9703 vs. 0.9471, p < 0.0001, patient-wise). Among the three subgroups of reviewers, non-radiologist physicians demonstrate the greatest improvement in diagnostic accuracy for brain CT interpretation with AI assistance compared to that without AI assistance. For board-certified radiologists, the diagnostic accuracy for brain CT interpretation is significantly higher with AI assistance than without AI assistance. For neuroradiologists, although brain CT interpretation with AI assistance results in a trend for higher diagnostic accuracy compared to that without AI assistance, the difference does not reach statistical significance. For the detection of AIH, brain CT interpretation with AI assistance results in better diagnostic performance than that without AI assistance, with the most significant improvement observed for non-radiologist physicians.

Park C, et al. Yonsei Med J. 2022 Jan;63(Suppl):S74-S83. https://doi.org/10.3349/ymj.2022.63.S74

This study aimed to investigate the relationships among parameters from dynamic contrast-enhanced (DCE) MRI, diffusion-weighted MRI (DWI), and ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) PET in patients with primary head-and-neck squamous cell carcinoma (HNSCC). A total of 34 patients with primary HNSCC underwent DCE-MRI, DWI, and ¹⁸F-FDG PET before treatment. The perfusion parameters (K_trans, K_transmax, K_ep, V_e, V_p, and AUC₆₀) from DCE-MRI and ADC (ADC_mean, ADC_min) values from DWI were calculated within the manually placed ROI around the main tumor. Standardized uptake value (SUV_max, SUV_mean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG = SUV_mean × MTV) were calculated with thresholds of 3.0 SUV. The associations between parameters were evaluated by Pearson correlation analysis. Significant correlations were identified between K_trans and K_ep (r = 0.631), K_trans and V_e (r = 0.603), K_trans and ADC_mean (r = 0.438), K_transmax and K_ep (r = 0.667), K_transmax and V_p (r = 0.351), V_e and AUC₆₀ (r = 0.364), V_e and ADC_mean (r = 0.590), and V_e and ADC_min (r = 0.361). ADC_min was reversely correlated with TLG (r = –0.347). Tumor volume was significantly associated with K_transmax (r = 0.348). The demonstrated relationships among parameters from DCE, DWI, and ¹⁸F-FDG PET suggest complex interactions among tumor biologic characteristics. Each diagnostic technique may provide complementary information for HNSCC.