New LungQ Software Can Analyze CF Airway Problems in Seconds, Thirona Says
Thirona announced it has developed a new software program that can quickly detect and quantify lung anomalies in cystic fibrosis (CF) patients, including those beyond the capacity of a doctor in a clinic.
The software is based on an artificial intelligence (AI) algorithm, called PRAGMA-AI, that is reported to allow for an analysis of CT scans that’s both fast — in a matter of seconds — and fully automated, so no human intervention is needed.
According to Thirona, a Dutch company specialized in AI, automated PRAGMA-AI could be used in both clinical trials and in clinical care to improve diagnosis, patient monitoring, and treatment planning.
“It allows for large scale investigation of CF lung disease … a crucial component in the evaluation of new expensive treatment options for patients suffering from CF,” Harm Tiddens, MD, PhD, co-developer of the first PRAGMA software version, said in a press release.
“CF patient registries, collecting information on the health status of CF patients, have already shown interest in the PRAGMA-AI method. They use patient information to create care guidelines, drive quality improvement, and to study CF treatments and outcomes,” Tiddens added.
An algorithm is basically a set of instructions designed to perform a specific task. AI algorithms can “learn” data, to quickly produce outputs when given new inputs.
Researchers at Erasmus University Medical Centre Rotterdam, working with the Telethon Kids Institute in Australia, originally created PRAGMA-CF, a software to quantify problems seen on CT scans, a standard diagnostic tool with CF. But this software’s use required highly trained professionals and was time-consuming.
Thirona’s new program is faster and user-friendly. While PRAGMA-CF would take up several hours to analyze CT scans, PRAGMA-AI is reported to perform the same tasks in “several seconds” and without human assistance.
PRAGMA-AI, part of Thirona’s LungQ software package, is said to quickly measure the exact extent of CF lung damage, like collapsed lung tissue or airway abnormalities, with high precision. Clinicians can use this information in making treatment decisions and monitoring disease progression.
PRAGMA-AI’s algorithm has been validated “on a large number of scans of CF patients, showing high diagnostic performance comparable to trained human analysts,” the company reported in its release. LungQ software is cleared for commercial use in both the U.S. and across the European Union.
Thirona also offers software packages for other diseases, including CAD4TB for diagnosing tuberculosis, and RetCAD for detecting anomalies due to age-related macular degeneration and diabetic retinopathy.
PRAGMA-AI software will be part of a next release of LungQ, Thirona stated, and available for clinical use in the U.S., EU, and Australia.