Arabella Asis Science 03.25.2024 02:Mar AM EDT

Researchers Train AI That Identifies COVID-19 From Lung Ultrasound Images

Researchers from Johns Hopkins University developed an automated detection tool that could help doctors identify COVID-19 based on lung ultrasound images.

The AI-powered tool can also be used in giving medical diagnostics for other pulmonary diseases, similar to how AI software can recognize faces in a crowd.

AI Helps Diagnose COVID-19, Pulmonary Diseases

Senior author Muyinatu Bell, the John C. Malone associate professor, shared that the new tool would be useful for emergency settings, especially when dealing with high patient caseloads. The AI model is capable of identifying B-lines in ultrasound lung images which indicates inflammation that is associated with complications.

The system will then analyze the combination of data and will accurately detect any abnormalities that relate to the virus.

"We are proving that with computer-generated datasets, we can still achieve a high degree of accuracy in detecting these COVID-19 features. The implications of this research extend beyond the current pandemic," Lingyi Zhao explained.

AI-Driven Medical Diagnostics to Aid Doctors on Emergency

The researchers acknowledged the potential of the new AI medical tool, citing that the ideal result is to produce wireless devices that can be used by patients at home to monitor the progression of COVID-19.

"An ideal use case would be wearable ultrasound patches that monitor fluid buildup and let patients know when they need a medication adjustment or when they need to see a doctor," co-author Tiffany Fong shared.

For now, the researchers are proving that computer-generated datasets can deliver highly accurate results in detecting COVID-19. The model currently uses simulated data to interpret real scans from patients with affected lungs as part of its training set.

"What we are doing here with AI tools is the next big frontier for point of care," Fong added.