BACKGROUND AND AIM: The aim of this study was to assess the diagnostic accuracy of e-CTA (Brainomix) in the automatic detection of large vessel occlusions (LVO) in anterior circulation stroke. METHODS: Of 487 CT angiographies (CTA) from patients with LVO stroke, 327 were used to train the algorithm while the remaining cases together with 140 negative CTAs were used to validate its performance against ground truth. Of these 301 cases, 144 were randomly selected and used for an additional comparative analysis against 4 raters. Sensitivity, specificity, positive and negative predictive value (PPV and NPV), accuracy and level of agreement with ground truth (Cohen’s Kappa) were determined and compared to the performance of a neuroradiologist, a radiology resident and two neurology residents. RESULTS: e-CTA had a sensitivity and specificity of 0.84 (0.77-0.89) and 0.96 (0.91-0.98) respectively for the detection of any LVO on the correct side in the whole validation cohort. This performance was identical in the comparative analysis subgroup and was within the range of physicians at different levels of expertise: 0.86-0.97 and 0.91-1.00, respectively. For the detection of proximal occlusions, it was 0.92 (0.84-0.96) and 0.98 (0.94-1.00) for the whole cohort and 0.93 (0.80-0.98) and 1.00 (0.95-1.00) for the comparative analysis, respectively for e-CTA. The range was 0.8-0.97 for sensitivity and 0.97-1.00 for specificity for the four physicians. CONCLUSIONS: The performance of e-CTA in detecting any LVO is comparable to less experienced physicians, but is similar to experienced physicians for detecting proximal LVOs.

Tools for medical image analysis have been developed to reduce the time needed to detect abnormalities and to provide more accurate results. Particularly, tools based on artificial intelligence and machine learning techniques have led to significant improvements in medical imaging interpretation in the last decade. Automatic evaluation of acute ischemic stroke in medical imaging is one of the fields that witnessed a major development. Commercially available products so far aim to identify (and quantify) the ischemic core, the ischemic penumbra, the site of arterial occlusion and the collateral flow but they are not (yet) intended as standalone diagnostic tools. Their use can be complementary; they are intended to support physicians' interpretation of medical images and hence standardise selection of patients for acute treatment. This review provides an introduction into the field of computer-aided diagnosis and focuses on the automatic analysis of non-contrast-enhanced computed tomography, computed tomography angiography and perfusion imaging. Future studies are necessary that allow the evaluation and comparison of different imaging strategies and post-processing algorithms during the diagnosis process in patients with suspected acute ischemic stroke; which may further facilitate the standardisation of treatment and stroke management.

Background and Purpose: Failure of early neurological improvement (fENI) despite successful mechanical thrombectomy in the anterior circulation is a clinically frequent occurrence. Purpose of this analysis was to define independent clinical, radiological, laboratory, or procedural predictors for fENI. Methods: Retrospective single-center analysis of patients treated for acute ischemic stroke in the anterior circulation ensuing successful mechanical thrombectomy between January 2014 and April 2019. Patients were compared according to fENI (equal or higher National Institutes of Health Stroke Scale) and ENI (lower National Institutes of Health Stroke Scale at discharge). Thirty-eight variables were examined in multivariable analysis for association with fENI. Results: Five hundred forty-nine out of 1146 patients experienced successful recanalization (modified Treatment in Cerebral Ischemia 2c-3). fENI occurred in 115/549 (20.9%) patients. Independent predictors of fENI were premorbid modified Rankin Scale (odds ratio [OR] per point [IC], 1.21 [1.00–1.46], P= 0.049), end-stage renal failure (OR [IC], 12.18 [2.01–73.63], P= 0.007), admission glucose (OR [IC], 1.018 [1.004–1.013] per mg/dL, P= 0.001), bridging IV lysis (OR [IC], 0.57 [0.35–0.93], P : 0.024), time from groin puncture to final recanalization (OR [IC], 1.004 [1.001–1.007] per minute, P= 0.015), general anesthesia during mechanical thrombectomy (OR, 2.41 [1.43–4.08], P <0.001), symptomatic intracranial hemorrhage (OR [CI], 6.81 [1.84–25.16], P= 0.004), and follow-up Alberta Stroke Program Early Computed Tomography Score (OR [IC], 0.76 [0.69–0.84] per point, P <0.001). In a secondary analysis, involvement of the regions internal capsule, M4 and M5 (motor cortex) were further independent predictors for fENI. Patients with ENI were more likely to experience a good outcome (modified Rankin Scale on day 90, 0–2: n=229/435 [52.8%] versus n=13/115 [11.3%]; P <0.001). Conclusions: The extent of infarction and the involvement of motor cortex and internal capsule as well as higher premorbid modified Rankin Scale, end-stage renal failure, high glucose level on admission, absence of bridging IV lysis, general anesthesia, and a longer therapy interval are presumably independent predictors for fENI in patients with successful mechanical thrombectomy.