Machine-learning algorithm in acute stroke: real-world experience

AIM: To assess the clinical performance of a commercially available machine learning (ML) algorithm in acute stroke.MATERIALS AND METHODS: CT and CT angiography (CTA) studies of 104 consecutive pa-tients (43 females, age range 19-93, median age 62) performed for suspected acute stroke at a single tertiary institution with real-time ML software analysis (RAPIDTM ASPECTS and CTA) were included. Studies were retrospectively reviewed independently by two neuroradiologists in a blinded manner. RESULTS: The cohort included 24 acute infarcts and 16 large vessel occlusions (LVO). RAPIDTM ASPECTS interpretation demonstrated high sensitivity (87.5%) and NPV (87.5%) but very poor specificity (30.9%) and PPV (30.9%) for detection of acute ischaemic parenchymal changes. There was a high percentage of false positives (51.1%). In cases of proven LVO, RAPIDTM ASPECTS showed good correlation with neuroradiologists' blinded independent interpretation, Pearson correlation coefficient = 0.96 (both readers), 0.63 (RAPIDTM vs reader 1), 0.69 (RAPIDTM vs reader 2).RAPIDTM CTA interpretation demonstrated high sensitivity (92.3%), specificity (85.3%), and negative predictive (NPV) (98.5%) with moderate positive predictive value (PPV) (52.2%) for detection of LVO (N=13). False positives accounted for 12.5% of cases, of which 27.3% were attributed to arterial stenosis.CONCLUSION: RAPIDTM CTA was robust and reliable in detection of LVO. Although demon-strating high sensitivity and NPV, RAPIDTM ASPECTS interpretation was associated with a high number of false positives, which decreased clinicians' confidence in the algorithm. However, in cases of proven LVO, RAPIDTM ASPECTS performed well and had good correlation with neu-roradiologists' blinded interpretation. (c) 2022 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Automated summary

This study aimed to evaluate the clinical performance of a commercially available machine learning algorithm in acute stroke. The results showed that the algorithm had high sensitivity and negative predictive value but poor specificity and positive predictive value for the detection of acute ischemic parenchymal changes.