Study shows more accurate way for Aortic Dissections through AI

Aortic dissections, in which the body’s main blood vessel’s inner layer rips, are an uncommon condition, occurring only in three out of every 100,000 individuals annually, but they have a disproportionately high fatality rate. According to estimates, 20% of individuals who experience acute aortic dissection pass away before they ever get to the hospital, and 50% of those who don’t receive treatment pass away within 24 hours.

Viz.ai has created an artificial intelligence program that seeks to identify potential aortic dissections as early as possible to at least partially offset these alarming figures. The precision of the technology was demonstrated in a study earlier this month.

Together with Avicenna, Viz created an artificial intelligence solution for Aortic dissection that was initiated last year. It contains a broader Viz Aortic suite component and contains instruments for three different aneurysm detection types. In addition to instantly notifying physicians when it detects a possible aneurysm or dissection, the software also enables them to study problematic CT images right on their cell phones and start assembling relevant team members using the Viz platform’s built-in communication features. The module was approved by the FDA in 2021.

Researchers used the deep learning AI to process more than 1,300 CT angiography images which have been gathered from around the United States in the study of the aortic dissection-spotting tool, the findings of which were presented at the VEITHSymposium.

When put to the test, the system successfully detected aortic dissections with 94% sensitivity and better than 97% specificity, while accurately excluding negative instances.

A positive predictive value of 80% was also recorded, indicating a significant probability that the scans the AI detected were confirmed cases of aortic dissection rather than just false positives. The algorithm’s negative predictive value was even greater, at more than 99%, reflecting how probable it is that a negative signal is indeed negative for the disease.

According to Viz, the study’s authors concluded that highly accurate AI-powered aortic dissection detection “may have a positive downstream effect on patient triage, leading to accelerated care coordination, earlier diagnosis, timely initiation of life-saving interventions, and better patient outcomes.”

The dissection-detecting algorithm is one of several AI tools Viz has created to assist expedite the triage of CT scans and other imaging data, and its encouraging findings come as the business continues to broaden its reach.

The company signed two new collaborations only last week. The first combines Illuminate, a maker of artificial intelligence (AI) tools and natural language processing software with Viz to identify high-risk patients in electronic medical records. These tools also help determine the severity of diseases and keep track of patients after diagnosis. According to the company, together they will improve care for those with aortic aneurysms by integrating Illuminate’s technology into the Viz Aortic module.

Viz is also collaborating with a clinical research organization to hasten the enrolment in clinical trials for investigations looking at novel neurovascular therapeutics. By scanning photos, the company’s software finds trial-eligible patients and automatically notifies research teams.