US-based Viz.ai has released new trial data that assessed its artificial intelligence (AI) powered neurovascular technology to detect and diagnose neurology pathologies including acute ischemic stroke, cryptogenic stroke, unruptured incidental aneurysm (UIA) and brain haemorrhage.
The Viz aneurysm algorithm was evaluated in a retrospective trial that examined head computed tomography (CT) scans. Across 963 scans, the algorithm showed a specificity of 96.8% and a negative predictive value of 97.2%. The software can be used to improve workflow and triage patients.
The user engagement with Viz.ai’s large vessel occlusions (LVO) detection software was evaluated in a multicentre, prospective, randomised clinical trial. The study found that there was an 11-minute reduction in door-to-groin puncture (DTG) time across all centres with Viz LVO implementation. DGT signifies the time required for a patient to reach the neuro-interventional radiology (IR) suite for mechanical thrombectomy, with a shorter time being associated with better clinical outcomes.
Another single-centre trial evaluated the impact of the software by measuring door-to-angiography time pre- and post-implementation of LVO software. The study showed that LVO software implementation was associated with shorter intervals between initial hospital contact and neurointerventional among transferred patients. The cleared by the FDA LVO software and is distributed in the US by Medtronic.
The evaluation of Viz.ai’s convolutional neural network (CNN), Viz ICH+, showed that the system can automatically quantify intracranial haemorrhage (ICH) and lateral ventricular (LV) volumes as well as midline shift (MLS) compared to the more time intensive, manual neuroradiologists segmentation. Viz.ai’s acute ischaemic stroke triage software showed a reduction in time-to-treatment and a reduction in unnecessary transfers during a two-year period when implemented in a large academic healthcare system.
The data from the abstracts demonstrate the use of Viz’s AI platform in improving workflows and disease detection. GlobalData forecasts AI to be a key driver of medical device innovation. The AI market generated approximately $93bn in sales in 2023, as per GlobalData analysis. Multiple companies, such as Roche and Amazon, have developed AI algorithms to improve workflows and disease diagnosis.
