Scientists identify a new AI tool that can be used in X -rays and other radiological reports that now accurately “reads” for prognosis.
Why do we need artificial intelligence when you have human intelligence?
The reasons are innumerable ranging from cost effectiveness to high accuracy rates. But specifically in the radiology sector, it is a well known fact that there is a shortage in the number of radiologists in the world. As a result these specialists are over burdened by the demand. In the last few years, there are many attempts in bringing artificial intelligence into healthcare to ease out the stress faced by our medical staff.
How does AI perform in comparison to radiologists?
Published in March 2023, a study shows that there is immense potential for AI to be incorporated into the field of Chest radiography. The AI tool was able to dragonize unremarkable, remarkable and critical cases with a 99% accuracy (1).
Some of the abnormalities picked by the AI tool in the chest x-rays.
The most surprising finding was just how sensitive this AI tool was for all kinds of chest disease. In fact, we could not find a single chest X-ray in our database where the algorithm made a major mistake.
Dr. Louis Lind Plesner, a coauthor of this paper
In another pragmatic study published in Feb 2023, the team from South Korea found that AI-led softwares were capable of improving the detection of nodules in the chest. It stands to be a good early detection method with a high accuracy (2).
Leaders in AI technology
AI in medicine is a not a very new advancement. Back in 2017, there was an introduction of an AI tool called Saykara (now acquired by Nuance). This tool enables book-keeping, communication and analysis of medical records. Since introductions of AI, many companies have developed their own versions and become leaders in this commercialization of this new technology.
1. GE Healthcare Edison Platform
2. Philips IntelliSpace Discover 3.0
3. Nuance PowerScribe 360 Clinical Guidance
All these companies have developed a user-friendly algorithm that uses the raw X-ray images to yield accurate results.
What happens to Radiologists?
While AI is a great technology to use for the fast and efficient analysis and delivery of reports, it cannot substitute traditional board certified radiologists. While the countless previous reports are used as knowledge base for training these AI tools, there is always a necessity for human intervention for the time being. AI might reduce the turnover time and enable hospitals to function with the short staffed crisis as of date but we have a long way to go for AI to have complete independence in this area of medicine.