Correlating Age and Hematoma Volume with Extent of Midline Shift in Acute Subdural Hematoma Patients: Validation of an Artificial Intelligence Tool for Volumetric Analysis.

OBJECTIVE: Decision for intervention in acute subdural hematoma patients is based on a combination of clinical and radiographic factors. Age has been suggested as a factor to be strongly considered when interpreting midline shift (MLS) and hematoma volume data for assessing critical clinical severity during operative intervention decisions for acute subdural hematoma patients. The objective of this study was to demonstrate the use of an automated volumetric analysis tool to measure hematoma volume and MLS and quantify their relationship with age. METHODS: A total of 1789 acute subdural hematoma patients were analyzed using qER-Quant software (Qure.ai, Mumbai, India) for MLS and hematoma volume measurements. Univariable and multivariable regressions analyzed association between MLS, hematoma volume, age, and MLS:hematoma volume ratio. RESULTS: In comparison to young patients (≤ 70 years), old patients (>70 years) had significantly higher average hematoma volume (old: 62.2 mL vs. young 46.8 mL, P < 0.0001), lower average MLS (old: 6.6 mm vs. young: 7.4 mm, P = 0.025), and lower average MLS:hematoma volume ratio (old: 0.11 mm/mL vs. young 0.15 mm/mL, P < 0.0001). Young patients had an average of 1.5 mm greater MLS for a given hematoma volume in comparison to old patients. With increasing age, the ratio between MLS and hematoma volume significantly decreases (P = 0.0002). CONCLUSIONS: Commercially available, automated, artificial intelligence (AI)-based tools may be used for obtaining quantitative radiographic measurement data in patients with acute subdural hematoma. Our quantitative results are consistent with the qualitative relationship previously established between age, hematoma volume, and MLS, which supports the validity of using AI-based tools for acute subdural hematoma volume estimation.

Artificial Intelligence-based automated CT brain interpretation to accelerate treatment for acute stroke in rural India: An interrupted time series study.

In resource-limited settings, timely treatment of acute stroke is dependent upon accurate diagnosis that draws on non-contrast computed tomography (NCCT) scans of the head. Artificial Intelligence (AI) based devices may be able to assist non-specialist physicians in NCCT interpretation, thereby enabling faster interventions for acute stroke patients in these settings. We evaluated the impact of an AI device by comparing the time to intervention (TTI) from NCCT imaging to significant intervention before (baseline) and after the deployment of AI, in patients diagnosed with stroke (ischemic or hemorrhagic) through a retrospective interrupted time series analysis at a rural hospital managed by non-specialists in India. Significant intervention was defined as thrombolysis or antiplatelet therapy in ischemic strokes, and mannitol for hemorrhagic strokes or mass effect. We also evaluated the diagnostic accuracy of the software using the teleradiologists' reports as ground truth. Impact analysis in a total of 174 stroke patients (72 in baseline and 102 after deployment) demonstrated a significant reduction of median TTI from 80 minutes (IQR: 56·8-139·5) during baseline to 58·50 (IQR: 30·3-118.2) minutes after AI deployment (Wilcoxon rank sum test-location shift: -21·0, 95% CI: -38·0, -7·0). Diagnostic accuracy evaluation in a total of 864 NCCT scans demonstrated the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) in detecting intracranial hemorrhage to be 0·89 (95% CI: 0·83-0·93), 0·99 (0·98-1·00), 0·96 (0·91-0·98) and 0·97 (0·96-0·98) respectively, and for infarct these were 0·84 (0·79-0·89), 0·81 (0·77-0·84), 0·58 (0·52-0·63), and 0·94 (0·92-0·96), respectively. AI-based NCCT interpretation supported faster abnormality detection with high accuracy, resulting in persons with acute stroke receiving significantly earlier treatment. Our results suggest that AI-based NCCT interpretation can potentially improve uptake of lifesaving interventions for acute stroke in resource-limited settings.