GE Healthcare and Lunit expand AI partnership to improve COVID-19 care

July 14, 2020
by Valerie Dimond, Contributing Reporter
Lunit, a medical artificial intelligence (AI) startup that develops advanced medical image analytics and data-driven imaging biomarkers using deep learning technology, has announced it will expand its collaboration with GE Healthcare. The company said in a press announcement that the effort should help make AI algorithms more accessible to clinicians, alleviate clinical strain and streamline workflows.

Lunit says this would be the first time a company brings commercially available AI products to an existing X-ray equipment manufacturer, such as GE Healthcare, and that the Lunit INSIGHT CXR will now be available to numerous global fixed, mobile, and R&F X-ray providers through GE’s Thoracic Care Suite at point-of-sale.

The Thoracic Care Suite provides eight Lunit INSIGHT CXR AI algorithms to quickly analyze chest x-ray findings and flag abnormalities to radiologists for review, including pneumonia, which is commonly linked to COVID-19, tuberculosis, lung nodules, and other radiological findings. The technology can also be used to identify high-risk COVID-19 cases while monitoring progression and regression of mild respiratory symptoms.

Lunit says the algorithm has a 97% to 99% accuracy rate (Area Under the Curve – AUC) and studies published in Radiology and JAMA Network Open, show that physicians have seen their performance increase up to 20%.

GE Healthcare announced another collaboration with the University of Oxford-led National Consortium of Intelligent Medical Imaging (NCIMI) in the U.K. to develop and test algorithms to aid in the diagnosis and management of COVID-19 pneumonia, since right now clinicians aren’t able to easily predict which patients who test positive for COVID-19 will deteriorate and require hospital admission for oxygen and possible ventilation. It’s also unclear which patients will suffer long-term consequences from the lung damage caused by COVID-19 pneumonia. The teams aim to develop algorithms incorporating data from thousands of patients medical imaging, laboratory, and clinical observations to provide a quicker diagnosis and prediction of progress and recovery.

"With approximately 1.44 billion chest X-ray exams taking place each year, radiologists are overwhelmed, especially as they may be looking for multiple indications per exam," Katelyn Nye, general manger of global Mobile Radiography and Artificial Intelligence, told HCB News. "Thoracic Care Suite provides much needed support by automatically analyzing images for the presence of pneumonia indicative of COVID-19 — as well as seven additional chest X-ray abnormalities — and provides much needed support to help quickly identify high-risk cases as well as monitor patients' respiratory symptoms."

The two companies are also working to improve GE’s Edison Open AI Orchestrator, an algorithm management solution that helps healthcare providers work more efficiently and effectively. Lunit says the technology is available through a stand-alone server or integrated into GE Healthcare’s Centricity Picture Archiving and Communication System (PACS) and Enterprise Imaging System (EIS). When complete, the new tech will reduce the complexity of implementing and managing multiple systems and algorithms while making it easy to adopt and explore AI.

“We have been applying our AI into various types of medical images. Among them, Lunit INSIGHT CXR is one of our major products that has been commercialized since a few years ago," said Brandon Suh, CEO of Lunit in a statement. "We will continue to push forward and cooperate with market leaders through extended partnerships and collaborations, increasing the number of global use cases of our AI.”