AI from Saudi Arabia helps diagnose post-Covid lung problems

A new computer-aided diagnostic tool could help to overcome some of the challenges of monitoring lung health after viral infection, scientists believe.

The method, developed by Saudi Arabia's King Abdullah University of Science and Technology (Kaust), is known as deep-lung parenchyma-enhancing (DLPE). It overlays artificial intelligence algorithms on top of standard chest imaging data, which reveals otherwise indiscernible visual features indicative of lung dysfunction.

Through DLPE augmentation, “radiologists can discover and analyse novel sub-visual lung lesions”, Kaust computer scientist and computational biologist Xin Gao said. “Analysis of these lesions could then help explain patients’ respiratory symptoms,” allowing for better disease management and treatment, he said.

Mr Xin said like other respiratory illnesses, Covid-19 can cause lasting harm to the lungs but doctors have struggled to visualise this damage. Conventional chest scans do not reliably detect signs of lung scarring and other pulmonary abnormalities. This makes it difficult to track the health and recovery of people with persistent breathing problems and other post-Covid complications.

Although the Kaust team developed DLPE primarily with post-Covid recovery in mind, they also tested the platform on chest scans taken from people with various other lung problems, including pneumonia, tuberculosis and lung cancer. The researchers showed how their tool could serve as a broad diagnostic aide for all lung diseases, empowering radiologists to, as Mr Xin put it, “see the unseen".

Mr Xin told The National: "Scientifically, it means that the method is not 'overfitted' to just Covid-19 cases, but has great generalisation power to other lung diseases. It means that clinicians now have an interpretable AI method that can cover a wide range of lung diseases, which will definitely help them."

He pointed out that the technology has already been used at the First Affiliated Hospital of Harbin Medical University in China for clinical diagnosis and a former version of has been used to help frontline radiologists at King Faisal Specialist Hospital in Riyadh.

Mr Xin and members of his structural and functional bioinformatics group and the Computational Bioscience Research Centre created the tool. Also involved were artificial intelligence researcher and current Kaust provost Lawrence Carin and clinical collaborators from Harbin Medical University.

The method first eliminates any anatomical features not associated with the lung parenchyma — the tissues involved in gas exchange serving as the main sites of Covid-19-induced damage. That means removing airways and blood vessels, then enhancing the pictures of what is left behind to expose lesions that might be missed without the computer’s help.

The researchers trained and checked their algorithms using computed tomography (CT) chest scans from thousands of people in hospital with Covid-19 in China. They refined the method with input from expert radiologists and then applied DLPE for dozens of Covid-19 survivors with lung problems, all of whom had required intensive care treatment.

In this way, Mr Xin and his colleagues demonstrated that the tool could reveal signs of pulmonary fibrosis in long-Covid sufferers, thus helping to account for shortness of breath, coughing and other lung problems. This was a diagnosis, he said, that would be impossible with standard CT image analytics.

“With DLPE, for the first time, we proved that long-term CT lesions can explain such symptoms,” Mr Xin said. “Thus, treatments for fibrosis may be very effective at addressing the long-term respiratory complications of Covid-19.”