Two-stage convolutional neural network for segmentation and detection of carotid web on CT angiography.

BACKGROUND: Carotid web (CaW) is a risk factor for ischemic stroke, mainly in young patients with stroke of undetermined etiology. Its detection is challenging, especially among non-experienced physicians. METHODS: We included patients with CaW from six international trials and registries of patients with acute ischemic stroke. Identification and manual segmentations of CaW were performed by three trained radiologists. We designed a two-stage segmentation strategy based on a convolutional neural network (CNN). At the first stage, the two carotid arteries were segmented using a U-shaped CNN. At the second stage, the segmentation of the CaW was first confined to the vicinity of the carotid arteries. Then, the carotid bifurcation region was localized by the proposed carotid bifurcation localization algorithm followed by another U-shaped CNN. A volume threshold based on the derived CaW manual segmentation statistics was then used to determine whether or not CaW was present. RESULTS: We included 58 patients (median (IQR) age 59 (50-75) years, 60% women). The Dice similarity coefficient and 95th percentile Hausdorff distance between manually segmented CaW and the algorithm segmented CaW were 63.20±19.03% and 1.19±0.9 mm, respectively. Using a volume threshold of 5 mm3, binary classification detection metrics for CaW on a single artery were as follows: accuracy: 92.2% (95% CI 87.93% to 96.55%), precision: 94.83% (95% CI 88.68% to 100.00%), sensitivity: 90.16% (95% CI 82.16% to 96.97%), specificity: 94.55% (95% CI 88.0% to 100.0%), F1 measure: 0.9244 (95% CI 0.8679 to 0.9692), area under the curve: 0.9235 (95%CI 0.8726 to 0.9688). CONCLUSIONS: The proposed two-stage method enables reliable segmentation and detection of CaW from head and neck CT angiography.

Influence of the interventionist's experience on outcomes of endovascular thrombectomy in acute ischemic stroke: results from the MR CLEAN Registry.

BACKGROUND: The relationship between the interventionist's experience and outcomes of endovascular thrombectomy (EVT) for acute ischemic stroke of the anterior circulation, is unclear. OBJECTIVE: To assess the effect of the interventionist's level of experience on clinical, imaging, and workflow outcomes. Secondly, to determine which of the three experience definitions is most strongly associated with these outcome measures. METHODS: We analysed data from 2700 patients, included in the MR CLEAN Registry. We defined interventionist's experience as the number of procedures performed in the year preceding the intervention (EXPfreq), total number of procedures performed (EXPno), and years of experience (EXPyears). Our outcomes were the baseline-adjusted National Institutes of Health Stroke Scale (NIHSS) score at 24-48 hours post-EVT, recanalization (extended Thrombolysis in Cerebral Infarction (eTICI) score ≥2B), and procedural duration. We used multilevel regression models with interventionists as random intercept. For EXPfreq and EXPno results were expressed per 10 procedures. RESULTS: Increased EXPfreq was associated with lower 24-48 hour NIHSS scores (adjusted (a)ß:-0.46, 95% CI -0.70 to -0.21). EXPno and EXPyears were not associated with short-term neurological outcomes. Increased EXPfreq and EXPno were both associated with recanalization (aOR=1.20, 95% CI 1.11 to 1.31 and aOR=1.08, 95% CI 1.04 to 1.12, respectively), and increased EXPfreq, EXPno, and EXPyears were all associated with shorter procedure times (aß:-3.08, 95% CI-4.32 to -1.84; aß:-1.34, 95% CI-1.84 to -0.85; and aß:-0.79, 95% CI-1.45 to -0.13, respectively). CONCLUSIONS: Higher levels of interventionist's experience are associated with better outcomes after EVT, in particular when experience is defined as the number of patients treated in the preceding year. Every 20 procedures more per year is associated with approximately one NIHSS score point decrease, an increased probability for recanalization (aOR=1.44), and a 6-minute shorter procedure time.

Benefit of successful reperfusion achieved by endovascular thrombectomy for patients with ischemic stroke and moderate pre-stroke disability (mRS 3): results from the MR CLEAN Registry.

BACKGROUND: Pre-stroke dependent patients (modified Rankin Scale score (mRS) ≥3) were excluded from most trials on endovascular treatment (EVT) for acute ischemic stroke (AIS) in the anterior circulation. Therefore, little evidence exists for EVT in those patients. We aimed to investigate the safety and benefit of EVT in pre-stroke patients with mRS score 3. METHODS: We used data from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic stroke in the Netherlands (MR CLEAN) Registry. All patients treated with EVT for anterior circulation AIS with pre-stroke mRS 3 were included. We assessed causes for dependence and compared patients with successful reperfusion (defined as expanded Thrombolysis in Cerebral Ischemia scale (eTICI) 2b-3) to patients without successful reperfusion. We used regression analyses with pre-specified adjustments. Our primary outcome was 90-day mRS 0-3 (functional improvement or return to baseline). RESULTS: A total of 192 patients were included, of whom 82 (43%) had eTICI <2b and 108 (56%) eTICI ≥2b. The median age was 80 years (IQR 73-87). Fifty-one of the 192 patients (27%) suffered from previous stroke and 36/192 (19%) had cardiopulmonary disease. Patients with eTICI ≥2b more often returned to their baseline functional state or improved (n=26 (26%) vs n=15 (19%); adjusted odds ratio (aOR) 2.91 (95% CI 1.08 to 7.82)) and had lower mortality rates (n=49 (49%) vs n=50 (64%); aOR 0.42 (95% CI 0.19 to 0.93)) compared with patients with eTICI <2b. CONCLUSIONS: Although patients with AIS with pre-stroke mRS 3 comprise a heterogenous group of disability causes, we observed improved outcomes when patients achieved successful reperfusion after EVT.

Efficacy of postmortem CT and tissue sampling in establishing the cause of death in clinical practice: a prospective observational study.

AIMS: The aim of this study is to evaluate whether agreement with autopsy-determined cause of death (COD) increases by use of postmortem CT (PMCT) or PMCT in combination with postmortem sampling (PMS), when compared with clinical assessment only. METHODS: This prospective observational study included deceased patients from the intensive care unit and internal medicine wards between October 2013 and August 2017. The primary outcome was percentage agreement on COD between the reference standard (autopsy) and the alternative postmortem examinations (clinical assessment vs PMCT or PMCT+PMS). In addition, the COD of patient groups with and without conventional autopsy were compared with respect to involved organ systems and pathologies. RESULTS: Of 730 eligible cases, 144 could be included for analysis: 63 underwent PCMT without autopsy and 81 underwent both PMCT and autopsy. Agreement with autopsy-determined COD was significantly higher for both PMCT with PMS (42/57, 74%), and PMCT alone (53/81, 65%) than for clinical assessment (40/81, 51%; p=0.007 and p=0.03, respectively). The difference in agreement between PMCT with PMS and PMCT alone was not significant (p=0.13). The group with autopsy had a significantly higher prevalence of circulatory system involvement and perfusion disorders, and a lower prevalence of pulmonary system involvement. CONCLUSION: PMCT and PMS confer additional diagnostic value in establishing the COD. Shortcomings in detecting vascular occlusions and perfusion disorders and susceptibility to pulmonary postmortem changes could in future be improved by additional techniques. Both PMCT and PMS are feasible in clinical practice and an alternative when autopsy cannot be performed.