AI medical imaging application developer Oxipit was granted CE Class IIb certification for ChestLink autonomous AI imaging suite. The application autonomously reports on chest X-rays featuring no abnormalities without any involvement from a radiologist. It is the first regulatory approved AI medical imaging application to perform diagnostics autonomously.
"ChestLink ushers in the era of AI autonomy in healthcare - something we have been promised by medical futurists and technology experts. It presents the first case where a medical diagnostic evaluation will be carried out solely by an artificial intelligence application. ChestLink showcases the future of healthcare diagnostics, where AI operates as an integral part of the clinical workflow", says CEO of Oxipit Gediminas Peksys.
According to Gediminas Peksys, ChestLink will help to address radiologist shortage, automating 15-40% of daily reporting workflow depending on the type of medical institution. This is especially relevant in a primary care setting, where a vast majority of chest X-ray studies feature no abnormalities. A similar autonomy framework can be adapted for large scale screening projects, such as the global tuberculosis effort.
ChestLink produces final reports for healthy patient X-ray studies where the application is highly confident the X-ray features no abnormalities. ChestLink aims to automate the 'radiologist invariant' part of chest X-ray workflow scope, where - including aspects of patient age, clinical context and varying radiologist's subjectivity - the study would appear normal to any given radiologist.
Prior to certification, ChestLink has been operating in a supervised reporting setting in multiple pilot locations for more than a year, processing more than 500.000 real-world chest X-ray images.
"The sensitivity metric of 99% has translated to zero clinically relevant errors at our deployment institutions during the application piloting stage", adds Gediminas Peksys.
For operational oversight ChestLink application provides an analytics page with real-time updates and daily summaries on what cases were autonomously reported on, allowing to quickly trace the steps of application decisions.
Prior to autonomous operations, ChestLink deployments start with a retrospective imaging audit. Retrospective analysis helps to identify what part of studies at the medical institution can be successfully automated. The operations then move into a supervised setting, where ChestLink reports are validated by the Oxipit medical team and radiologists at the medical institution. Only after completing the initial stages, the application can start to report autonomously.
The CE mark paves the way for clinical ChestLink deployments in 32 European countries. Fully autonomous ChestLink operations in a clinical setting are expected to begin in early 2023.