Fixed-site diagnostic imaging service RadNet and its artificial intelligence-based software subsidiary Quantib have announced that it has received FDA clearances for its DeepHealth Saige-DX mammography and Quantib Prostate 2.0 MRI artificial intelligence (AI) algorithms.
“We are very proud that two of our subsidiaries, DeepHealth and Quantib, have received FDA clearance for their flagship products,” said Dr. Howard Berger, president and CEO of RadNet, in a press release. “Artificial intelligence will have a transforming impact on radiology and cancer care, and we are committed to delivering those advances to patients and healthcare providers. These recent approvals will help us drive improved patient outcomes while lowering costs. We believe that these AI tools will play an important role in the early detection and diagnosis of cancer, resulting in improved survival rates and better patient outcomes.”
Saige-Dx for mammography
For mammography images, the Saige-Dx algorithm is the next generation of the company’s FDA-approved Saige-Q software, which provides work-list triage. The newly approved Saige-Dx provides radiologists with a cancer detection tool to help them more accurately diagnose both the presence or absence of cancer using the mammogram image. The tool leverages the AI technology of another RadNet subsidiary, DeepHealth to identifies suspicious lesions in mammograms and to assign lesions found with a suspicion level to each finding and to the entire case. According to the company, this capability can detect and diagnose breast cancer earlier while reducing unnecessary recalls.
“In our study to support FDA clearance, all 18 radiologists who participated exhibited improved performance when using Saige-Dx, resulting in the largest increase in performance across all mammography AI products on the market to date. The improved performance was reflected in both an increase in the percentage of the cancers detected and a lower false positive rate when using Saige-Dx,” noted Bill Lotter, Ph.D., CTO, and co-founder of DeepHealth.
The tool has also shown its ability to provide diagnosis of the presence of cancer well in advance of current methods.
“Our highly accurate AI, as is described in a Nature Medicine article last year, showed the ability to find cancer one to two years earlier than fellowship-trained breast imaging radiologists. The approval of Saige-Dx represents another step forward in advancing cancer care,” added DeepHealth CEO and Co-Founder Gregory Sorensen, M.D.
Quantib Prostate 2.0 MRI
For prostate screening, RadNet’s Quantib Prostate 2.0 MRI improves prosate screening workflow and can be accessed directly via a radiologist’s workstation. The suite of tools provides AI-based segmentations and volumetry, PSA density calculation, precise registration and movement correction, one-click segmentation of lesion candidates, PI-RADS scoring support, and standardized reporting to facilitate communication of results.
The FDA provided 510k special clearance for this tool based on a major upgrade to Quantib from its previously release version 1.3, with Quantib Prostate 2.0 MRI now including automatic prostate zone segmentation (in addition to prostate gland segmentation) and automated initiation of localization of lesions on the PI-RADS sector map.
According to Arthur Post Uiterweer, CEO of Quantib: We seek to continuously improve our software to support radiologists and urologists in the best way possible, and we deliver updated solutions to our customers as quickly as possible.”