Selective focus of CT scan of Chest or lung axial mip view of lung infection covid-19 with ground glass opacity .
Credit: mr.suphachai praserdumrongchai/Getty Images

Outpatient diagnostic imaging company RadNet announced today that it has launched a licensing deal with Google Health to develop an artificial intelligence (AI)-based lung nodule malignancy detection technology to provide early and accurate diagnosis of lung cancer. The collaboration, via RadNet’s lung AI subsidiary Aidence, will seek to commercialize Google’s research model for lung nodule malignancy prediction on computed tomography (CT) imaging.

Most lung nodules are non-cancerous. However, current methods for distinguishing malignant versus non-malignant lung nodules is often a time-consuming endeavor involving multiple CT scans and invasive procedures. Low-dose CT as a screening method has begun to gain favor in the US and in Europe. The NELSON randomized clinical trial result examining the use of CT scans to screen for lung cancer is high-risk patient, published in 2020 in the New England Journal of Medicine, showed that it reduced lung cancer mortality by as much as 24% for men and 33% for women.

“One of the most exciting developments in contemporary population healthcare is the early detection of lung cancer. Unfortunately, the reality that most such nodules will be benign represents a real challenge that cries out for a technological solution. Artificial intelligence is one such solution,” said Raymond Osarogiagbon, chief scientist, Baptist Memorial Health Care Corporation and director of the Multidisciplinary Thoracic Oncology Program, Baptist Cancer Center in Memphis, Tennessee.

According to a 2018 paper published in Nature, the use of deep learning methods, a form of AI, can aid in the risk scoring of lung nodule malignancy. In that study, scientists affiliated with Google Health presented a highly accurate model for malignancy classification, which consistently matched the performance of experienced radiologists. Aidence scientists have shown similar success in these types of diagnoses, with its own 2021 clinical study of 300 cases comparing its AI-driven approach showing a robust performance compared with radiologists.

The Google Health, Aidence work will center on leveraging Google’s expertise in this area with Aidence further refining the model with an eye toward bringing to market a clinical tool that complies with data privacy and regulatory requirements.

“Our mission at Aidence is to give lung cancer patients a fighting chance. This strategic partnership with Google Health allows us to accelerate and expand our efforts toward achieving it,” said Mark-Jan Harte, Aidence co-founder and CEO. “We are enthusiastic about working on a powerful deep learning model for lung nodule malignancy prediction based on the work of the Aidence and Google teams, as well as making sure that all the other requirements that contribute to the successful deployment of AI in clinical practice are in place, like clinical validation, certification and integration into the clinical workflow.”

Aidence already has demonstrated success in providing AI-based diagnostic and screening tools. It’s application, Veye Lung Nodules, is currently running in over 80 routine practice and lung cancer screening sites in hospitals and clinics in the EU. Aidence’s parent company RadNet operates 349 freestanding, fixed-site diagnostic imaging services and related information technology solutions in U.S. markets including Arizona, California, Delaware, Florida, Maryland, New Jersey, and New York.

Also of Interest