FDA clears behold.ai's red dot algorithm for radiology triage in pneumothorax

The FDA cleared behold.ai's red dot cognitive computing technology to assist radiologists in providing "instant triage" in pneumothorax cases, the company announced Thursday, adding that it plans to launch the platform in the US later this year. "This technology augments the expertise of radiologists to enable them to report with greater clinical accuracy, faster and more safely than they could before," helping to "deliver greater performance in radiology reporting at a fraction of the price of outsourced reporting," behold.ai said.

According to the company, the red dot prioritisation platform can sort images into normal and abnormal categories in less than 30 seconds post image acquisition. It noted that the platform was developed using more than 30,000 example images, which were reviewed and reported by radiologists.

Simon Rasalingham, CEO at behold.ai, remarked that "our highly-trained and experienced UK consultant radiologists have continually reviewed, annotated and reported on patient images to help train the algorithm correctly. In the US, three independent board-certified radiologists checked whether the results were correct against more than 800 cases from the UK and US to validate the performance of the algorithm." He added that "in the UK, our momentum continues with the roll out of the red dot solution across the NHS."

The company said that in January, data presented at the British Thoracic Oncology Group (BTOG) conference showed the potential of the algorithm for "rapid detection of cancerous abnormalities" on chest x-rays. The technology was also showcased at the Artificial Intelligence in Clinical Radiology & Medical Imaging conference last month. Meanwhile, behold.ai indicated that it is in the final stages of obtaining CE mark Class IIa certification for the red dot algorithm.

Did you like this article?