Evaluating artificial intelligence software for delineating hemorrhage extent on CT brain imaging in stroke: AI delineation of ICH on CT.

BACKGROUND: The extent and distribution of intracranial hemorrhage (ICH) directly affects clinical management. Artificial intelligence (AI) software can detect and may delineate ICH extent on brain CT. We evaluated e-ASPECTS software (Brainomix Ltd.) performance for ICH delineation. METHODS: We qualitatively assessed software delineation of ICH on CT using patients from six stroke trials. We assessed hemorrhage delineation in five compartments: lobar, deep, posterior fossa, intraventricular, extra-axial. We categorized delineation as excellent, good, moderate, or poor. We assessed quality of software delineation with number of affected compartments in univariate analysis (Kruskall-Wallis test) and ICH location using logistic regression (dependent variable: dichotomous delineation categories 'excellent-good' versus 'moderate-poor'), and report odds ratios (OR) and 95 % confidence intervals (95 %CI). RESULTS: From 651 patients with ICH (median age 75 years, 53 % male), we included 628 with assessable CTs. Software delineation of ICH extent was 'excellent' in 189/628 (30 %), 'good' in 255/628 (41 %), 'moderate' in 127/628 (20 %), and 'poor' in 57/628 cases (9 %). The quality of software delineation of ICH was better when fewer compartments were affected (Z = 3.61-6.27; p = 0.0063). Software delineation of ICH extent was more likely to be 'excellent-good' quality when lobar alone (OR = 1.56, 95 %CI = 0.97-2.53) but 'moderate-poor' with any intraventricular (OR = 0.56, 95 %CI = 0.39-0.81, p = 0.002) or any extra-axial (OR = 0.41, 95 %CI = 0.27-0.62, p<0.001) extension. CONCLUSIONS: Delineation of ICH extent on stroke CT scans by AI software was excellent or good in 71 % of cases but was more likely to over- or under-estimate extent when ICH was either more extensive, intraventricular, or extra-axial.

Looking back to look forward: Surviving COVID-19 and the future of the NHS.

The COVID-19 pandemic affected healthcare systems worldwide, including the National Health Service (NHS). It drastically changed the practice and delivery of healthcare and laid bare longstanding structural flaws. It also brought a time of innovation and digitalisation and renewed appreciation of the role of public health. This paper offers a thematic summary of a debate held in December 2021 by the University of Edinburgh School of Medicine. It featured a multi-specialty panel of doctors and patient representative discussing the likely impact of the pandemic on the future of NHS. It serves as a reflection point on the pressures the NHS has faced since and their likely genesis at a time when the impact of the pandemic on staff risks being forgotten.

h-Index, Journal Citation Indicator, and Other Impact Factors in Neurosurgical Publications: Is There a Cost Factor That Determines the Quality?

OBJECTIVE: There has been an increase in number of Neurosurgical publications, including open access (OA), in recent years. We aimed to compare journals' performance and the relationship to submission fees incurred in publication. METHODS: We identified 53 journals issuing neurosurgery-related work. Quantitative analysis from various search engines involved obtaining h-index, Journal Citation Indicator (JCI), and other metrics such as Immediacy Index and 5-year impact factor utilising Journal Citation Reports. OA fees and individual subscription fees were collected. Correlations were produced using Spearman rho (ρ) (P < 0.05). RESULTS: Median h-index for 53 journals was 54 (range: 0-292), with JCI median reported as 0.785 (range: 0-2.45). Median Immediacy Index was 0.797 (range: 0-4.076), and median for 5-year impact factor was 2.76 (range: 0-12.704). There was a very strong positive correlation between JCI and Immediacy Index, JCI and 5-year impact factor, and 5-year impact factor and Immediacy Index (ρ > 0.7, P < 0.05). It is unclear whether there was any correlation between the indices and the OA costs and subscription costs for personal usage (P > 0.05). CONCLUSIONS: Larger costs incurred for OA fees and subscription costs for personal use do not clearly reflect on the journals' performance, as quantified by using various indices. There appears to be a strong association with performance across the journals' metrics. It would be beneficial to include learning about bibliometric indices' impact for research publications in medical education training to maximize the quality of the scientific work produced and increase the visibility of the information produced. The potential full movement to exclusively OA journals would create a significant barrier for junior researchers, small institutions, and full time-trainee physicians with limited funding available. This study suggests the need for a robust measurement of journals' output and the quality of the work produced.

External Validation of e-ASPECTS Software for Interpreting Brain CT in Stroke.

OBJECTIVE: The purpose of this study was to test e-ASPECTS software in patients with stroke. Marketed as a decision-support tool, e-ASPECTS may detect features of ischemia or hemorrhage on computed tomography (CT) imaging and quantify ischemic extent using Alberta Stroke Program Early CT Score (ASPECTS). METHODS: Using CT from 9 stroke studies, we compared software with masked experts. As per indications for software use, we assessed e-ASPECTS results for patients with/without middle cerebral artery (MCA) ischemia but no other cause of stroke. In an analysis outside the intended use of the software, we enriched our dataset with non-MCA ischemia, hemorrhage, and mimics to simulate a representative "front door" hospital population. With final diagnosis as the reference standard, we tested the diagnostic accuracy of e-ASPECTS for identifying stroke features (ischemia, hyperattenuated arteries, and hemorrhage) in the representative population. RESULTS: We included 4,100 patients (51% women, median age = 78 years, National Institutes of Health Stroke Scale [NIHSS] = 10, onset to scan = 2.5 hours). Final diagnosis was ischemia (78%), hemorrhage (14%), or mimic (8%). From 3,035 CTs with expert-rated ASPECTS, most (2084/3035, 69%) e-ASPECTS results were within one point of experts. In the representative population, the diagnostic accuracy of e-ASPECTS was 71% (95% confidence interval [CI] = 70-72%) for detecting ischemic features, 85% (83-86%) for hemorrhage. Software identified more false positive ischemia (12% vs 2%) and hemorrhage (14% vs <1%) than experts. INTERPRETATION: On independent testing, e-ASPECTS provided moderate agreement with experts and overcalled stroke features. Therefore, future prospective trials testing impacts of artificial intelligence (AI) software on patient care and outcome are required before widespread implementation of stroke decision-support software. ANN NEUROL 2022;92:943-957.

Neurosurgical Publication-Should We Publish at Any Cost? An In-Depth Analysis of Costs Incurred in Publication.

OBJECTIVE: With the recent paradigm shift in neurosurgical publications, open access (OA) publishing is burgeoning along with traditional publishing methods. We aimed to explore costs of publication across 53 journals. METHODS: We identified 53 journals publishing neurosurgical work. Journal type, submission and open access charges, color print fees, impact indicators, publisher, and subscription prices were obtained from journal and publisher websites. Costs were unified in U.S. dollars. Mean prices per journal were used to equilibrate membership and subscription discounts. Correlations were performed using Spearman ρ (P < 0.05). RESULTS: Of 53 journals, 12 were OA only, 40 were hybrid, and 1 was traditional. Submission costs were provided by 22 and 43 journals, respectively, by the end of phase 1 and 2 (prices always for phase 2: 26 free of charge, 4 <$500, and 1 <$1000). Median OA charge was $3286 (49 journals; range, $0-$7827). Of 53 journals, 36 did not list print fees for color figures (29 in phase 2). Median fee estimate per figure was $422 (range, $25-$1060). Median personal subscription for 1 year was $344 (range, $60-$1158; 48 journals). Median institutional subscription for 1 year was $2082 (range, $38-$5510; 34 journals). There was a mild positive correlation between Journal Impact Factor and OA fees (ρ = 0.287, P = 0.046). CONCLUSIONS: The lack of easily accessible information about neurosurgical publications, such as submission costs or OA charges, creates an unnecessary hurdle and should be remedied. Publishing in neurosurgery should be a positive learning experience, and cost should not be a limiting factor.