• AI-powered 'ischemia or not' diagnostic algorithm performance improved
• Sensitivity increased from 85% to 95%, specificity from 78% to 80%
• Continuous innovation efforts to progress device development and enhance algorithms

EMVision Medical Devices Limited (ASX:EMV) has provided a progress update on its continuous innovation efforts, which have generated enhanced AI-powered 'ischemia or not' diagnostic performance. In the previous 'EMView' pre-validation trial, the AI-powered algorithm achieved a sensitivity of 85% and specificity of 78%. As part of the continuous innovation efforts, the algorithm has been re-trained using cleaned training data and re-evaluated. In this dataset, the algorithm's performance has improved, with a sensitivity of 95% and specificity of 80%. A limited sensitivity analysis showed the algorithm outperformed non-contrast computed tomography (NCCT) imaging and computed tomography perfusion (CTP) / computed tomography angiography (CTA) in detecting ischemic cases. While the data is encouraging, the company notes that due to the study design and limited sample size, the results do not yet allow for statistically significant or generalizable conclusions. EMVision is implementing a cost-effective strategy for continued device innovation and enhancement during the pivotal (validation) trial via a parallel Continuous Innovation Study, which will utilize additional patient scans at multiple sites in Australia. This study data will be used to progress the development of additional device features, scale the training library for EMVision's diagnostic AI algorithms, and potentially extend indications by including patients with traumatic brain injury.

EMVision is implementing a cost-effective strategy for continued device innovation and enhancement during the pivotal (validation) trial via a parallel Continuous Innovation Study. Additional patients will be scanned at multiple sites in Australia, and this study data will be used to progress the development of additional device features, scale the training library for EMVision's diagnostic AI algorithms, and potentially extend indications by the enrolment of patients with traumatic brain injury.