BEIJING, March 9 (Xinhua) -- Chinese researchers have developed an artificial intelligence (AI) assisted computed tomography (CT) analysis system in identifying patients with COVID-19 or common viral pneumonia, Science and Technology Daily reported Monday.
The AI-assisted CT image comprehensive analysis system was jointly developed by the AI-tech professional committee on tumor treatment under the China Anti-Cancer Association (CACA) and National Supercomputer Center in Tianjin.
It was developed with the target of offering a more accurate distinction of a patient with COVID-19 or common viral pneumonia amid the intensive clinical demands in the anti-epidemic fight.
The system has entered trial operation in the National Supercomputer Center in Tianjin, which develops and operates China's first petaflop supercomputer the Tianhe-1.
It will contribute helpful references to the CT image judgment on COVID-19 patients, according to Xu Bo, head of the research team and chairman of the AI-tech professional committee on tumor treatment under CACA.
The conventional CT image is helpful in the early screening of suspected cases with the coronavirus, but images of various viral pneumonia are quite similar. It is extremely difficult for doctors to give a diagnosis from visual screenings alone.
The AI-assisted CT image comprehensive analysis system generates a power of multiple technologies, including big data and AI deep learning. It will enhance capabilities in clinical judgment on COVID-19 patients, according to Xu.
"To enable the new AI-assisted system to conduct the deep learning, we must first have a training set with mass data," Xu said, adding that the AI-tech professional committee under the CACA exploits the advantages of mobilizing nationwide medical institutions to collect CT images of COVID-19 patients.
Then, the joint research team trained the system to learn and analyze the typical characteristics and their locations on those CT images, such as the lung consolidation and ground-glass opacity.
"These characteristics and locations are of critical importance for the modeling and learning, and the system's final precise judgment," said Xu.
National guidelines recommend CT scans as the key marker for diagnosing COVID-19.
Chinese scientific research institutions and high-tech companies have been ramping up efforts to develop AI systems to speed up the processing of CT scans of suspected COVID-19 cases.
On March 2, China's AI firm iFlytek announced it had jointly developed an AI-based COVID-19 diagnosis platform with the Chinese Academy of Sciences.
According to iFlytek, the system can read and analyze a patient's CT scans within three seconds. Deployed at a hospital in Hefei, Anhui Province, the system so far has identified all confirmed cases and has a recall rate of 90 percent in lesion detection.
It can analyze key image features such as the morphology, range and density of the lesion and present a dynamic 4D contrast of whole lung lesions, providing an accurate and efficient reference for COVID-19 diagnosis.
