ROME, March 23 (Xinhua) -- A university hospital in Rome has started to use a new diagnostic tool that uses artificial intelligence (AI) to help quickly diagnose patients with coronavirus.
The so-called "Infer-read" system is based on technology developed by a Chinese company called InferVision and adapted for use in Italy. Rome's Polyclinic Bio-Medical University purchased the license for the system and InferVision engineers based in Germany traveled to Rome to train staff on its use.
The system uses around 600 images of a patient's lungs taken through the use of standard computerized tomography scans to quickly determine which patients may have been infected by the coronavirus and which are uninfected.
Once the images are loaded into the system, it produces a result in around 20 seconds. Traditional diagnosis for coronavirus takes several hours if it is processed on-site, and up to three days if it must be sent away for processing.
"This new kind of system doesn't do the job on its own, but it can serve as a very important tool to help a skilled radiologist," said Bruno Beomonte Zobel, coordinator of the Imaging Center at the Polyclinic Bio-Medical University, where the system is housed.
Beomonte Zobel, who is also a professor at the university's diagnostic imaging and radiotherapy department, told Xinhua that the process relies on artificial intelligence to "learn" the characteristics of infected lungs. It is only applicable in areas where there are thousands of cases of viral infections for the system to understand how to recognize them.
It is not necessary that individual hospitals have the system on hand since they can upload the images from the computerized tomography scans -- a widely available technology, and send them to be analyzed. Beomonte Zobel said the system in Rome is capable of processing "a few hundred" cases a day, though it has yet to be used to full capacity.
Though the system reportedly boasts a 98.5-percent accuracy rate, Beomonte Zobel said its main use is to help radiologists decide which examinations to focus on.
"If we get the images for, let's say, 500 suspected cases of coronavirus, the system will be very good at determining that only 200 of them merit close inspection by a radiologist," he said. "That is a big time saver."
According to Luca Bianchi, chief technology officer for Neosperience, a technological innovation firm promoting the use of artificial intelligence to help curb the spread of coronavirus, the promising new technology has great potential to play an important role in the fight against the virus. But he said some challenges to its use remain.
"The system can interpret results quickly, but the computerized tomography scans still take time and after each one, the machine has to be thoroughly sanitized to assure that if the first patient is infected by the coronavirus, it will not be passed onto the next patient using the machine," Bianchi said in an interview.
