AI quality assurance study

A study by researchers at Mahajan Imaging and was conducted to evaluate the performance of diagnostic AI tools for quality assurance. The study found that nearly 1% of chest radiographs included clinically significant diagnostic mistakes, which were identified by Oxipit Quality AI software. The study was presented at the ECR2023 annual meeting earlier in March.

A retrospective sample of 3.892 anonymised chest X-ray studies was selected for the research. The studies were then processed by Oxipit Quality AI software.

Oxipit Quality software analyzes chest X-ray images and corresponding radiologist reports. Initially the software processes CXR images to identify any radiological findings. It then uses natural language processing to check for findings not described in the radiologist report.

Out of 3.892 studies, Oxipit software flagged 600 studies as potentially having unreported findings. 34 flagged findings (0.87%) were deemed clinically significant. These include 20 studies with nodular opacities, 11 containing consolidation, and one study each of pneumothorax, malposition of central venous line and pleural effusion.

“While this is a relatively low percentage, it still represents a significant number of potential errors that could be caught with the use of an AI-based tool. If you look at the scale of the operation of our institution, in which we do over 300,000 exams a year, that means 3,000 patients who would have a change in their treatment outcomes. Even such small numbers could change the lives of so many families”, – says Chief Medical Officer Dr Vasantha Venugopal, who presented the study.

“In previous clinical validation studies, we have recorded clinically significant missed findings rates from 0.18% to 0.92%. Operating in near-real-time, Oxipit Quality offers an user-friendly framework to eliminate these mistakes before any treatment decisions are taken”, – says Chief Medical Officer at Oxipit Naglis Ramanauskas.

Oxipit Quality has already processed more than 300.000 studies in a clinical setting. Oxipit Quality suite was recently expanded to support quality assurance for MSK X-rays, MG images and CT scans.