US Food and Drug Administration (FDA) granted Aidoc, the Tel Aviv startup specializing in the design of artificial intelligence (AI) driven radiology solutions its 6th regulatory clearance last Wednesday (21 October). The cleared algorithm will flag and communicate incidental pulmonary embolism (PE) and it’s expected to be deployed alongside Aidoc’s five other FDA approved tools to triage and notify Intracranial Haemorrhage, C-Spine Fractures, Large Vessel Occlusion and Intra-Abdominal Free Gas.

In the press release, Aidoc noted flagging of incidental, critical findings can be a huge technical challenge because of the many imaging protocols in use and low incidences of cases. As such, to accurately prioritize incidental finding of critical conditions becomes a breakthrough that AI can bring to the radiologist workflow.

“There’s a reason why most AI triage solutions don’t focus on incidental findings,” Michael Braginsky, Aidoc’s Chief Technology Officer says. “Because the prevalence of incidental findings is relatively low, the specificity of the AI must be especially high, otherwise false positive rate will be excessive and user adoption will be negatively impacted. In addition, an incidental PE algorithm detects PE in non-dedicated exams, where contrast is by definition suboptimal, and there’s an extremely high variability of protocols which challenges the AI even further. It was a scientific breakthrough that our team achieved that made this possible”.

Aidoc’s Vice-President in Marketing Ariella Shoham told AIMed via email this new approval is “specifically exciting” for the company because of its uniqueness. “No other AI has been cleared to flag incidental findings. This is a different clearance to the one we already have for ‘non-incidental’ PE”.

Shoham cited PE is the second-most common cause of death in cancer patients, for which the vast majority of all fatalities are only found after death. Early diagnosis and treatment significantly improve outcomes and saves lives. Studies suggesting an incidental prevalence of PE as high as 2.6% further emphasize the need for a solution that can ensure that these incidental cases are addressed in a timely manner”.

On top of which, Aidoc also briefly mentioned the decision made by the US Centers for Medicare and Medicaid Services (CMS) last month, in providing new technology add-on payment (NTAP) status for a software which detects stroke on CT scans to be extended to Aidoc’s AI algorithm for detecting and triaging Large Vessel Occlusion. The company is also  working to submit its other AI technology for possible inclusion in the NTAP reimbursement program.

In fact, Aidoc’s Chief Executive Officer Elad Walach had repeatedly expressed his excitement towards AI reimbursement in several recent blog posts. He wrote value is probably the main driver for CMS to reimburse the use of AI in clinical setting and AI had indeed, demonstrated its value by working with radiologists to expedite formal diagnosis, treatment, and reduce length of hospital stay. “It’s empowering to see that the CMS now recognizes this value and I look forward to additional reimbursement in the future as AI based triage becomes standard of care for additional pathologies”.

*

Author Bio

Hazel Tang A science writer with data background and an interest in the current affair, culture, and arts; a no-med from an (almost) all-med family. Follow on Twitter.