NEW YORK, BRONX (ORDO News) — In a groundbreaking study published in the journal Radiology , radiologists outperformed artificial intelligence (AI) in accurately identifying three common lung diseases from chest X-rays.
The study, conducted by a team led by Dr. Louis L. Plesner, MD, PhD, of Herlev and Gentofte Hospital in Copenhagen, Denmark, compared the performance of four commercial artificial intelligence tools and 72 radiologists in interpreting more than 2,000 chest x-rays.
Chest radiography is a widely used diagnostic tool in medicine. However, significant training and experience is required to correctly interpret images.
Although artificial intelligence tools have been developed to assist radiologists in making diagnoses, their clinical application is still in its early stages. Dr. Plesner emphasizes the need to further test AI tools in real-world clinical scenarios to determine their true diagnostic accuracy.
The study analyzed 2,040 consecutive adult chest X-rays taken over a two-year period at four Danish hospitals in 2020. The mean age of the patients was 72 years, and 32.8% of radiographs had at least one target finding related to three common lung diseases: airspace pathology, pneumothorax, and pleural effusion.
The results of the study showed that radiologists outperform artificial intelligence tools in accurately determining the presence and absence of these lung diseases. While AI tools were sensitive to detecting diseases, they also produced more false positives, making them less reliable for offline diagnosis. However, artificial intelligence tools can still be useful in obtaining second opinions to assist radiologists in their interpretation.
Dr. Plesner’s research sheds light on the current limitations of artificial intelligence in radiology diagnostics. It highlights the need for further research and testing to improve their diagnostic accuracy in real-world clinical settings. Dr. Plesner states, “AI tools can help radiologists interpret chest X-rays, but their diagnostic accuracy in real-world settings remains unclear.”
Scientist Quotes
– “Chest X-rays are a common diagnostic tool, but significant training and experience are required to correctly interpret the studies.” – Dr. Louis L. Plesner, Principal Investigator.
“Although artificial intelligence tools are increasingly being validated for use in radiology departments, there is an unmet need to further validate them in real-world clinical scenarios.” – Dr. Louis L. Plesner, Principal Investigator Conclusion.
This study highlights the superiority of radiologists over artificial tools accuracy in detecting common lung diseases from chest X-rays.Although artificial intelligence tools have demonstrated sensitivity in detecting these diseases, the high rate of false positives raises doubts about their reliability for offline diagnosis.
However, they can serve as a valuable tool for providing second opinions to assist radiologists. Further research and testing is needed to improve the diagnostic accuracy of AI tools in real-world clinical settings.
—
Online:
News agencies contributed to this report, edited and published by ORDO News editors.
Contact us: [email protected]
Our Standards, Terms of Use: Standard Terms And Conditions.
To eliminate any confusion arising from different time zones and daylight saving changes, all times displayed on our platforms are in Coordinated Universal Time (UTC).