AI-generated X-ray images fooled medical experts and improved osteoarthritis classification

The sharing of medical data between laboratories and medical experts is important for medical research. However, data sharing is often complex enough and sometimes even impossible due to strict EU data regulation laws. Researchers from the Digital Health Intelligence Laboratory at the University of Jyväskylä investigated the problem and developed an artificial neural network that creates synthetic X-ray images that can fool even medical experts.

A group of researchers from the AI ​​Hub Central Finland project at the University of Jyväskylä has developed an AI-based method to create synthetic X-ray images of the knee to replace or supplement real X-ray images in the classification of osteoarthritis of the knee.

The researchers used synthetically generated x-ray images to supplement a dataset of real x-ray images from the osteoarthritis study. The authenticity of the images was then assessed with specialists from the Central Finland Healthcare District.

Medical experts were asked to rate the severity of osteoarthritis, unaware that the dataset included synthetic images. In the second phase, the experts tried to identify authentic and synthetic images. The results showed that even medical experts were, on average, unlikely to distinguish between real and synthetic x-ray images.

“The use of synthetic data is not subject to the same data protection rules as real data. The use of synthetic data can facilitate collaboration between, for example, research groups, companies and educational institutions,” says Sami Äyrämö, head of the Digital Health Intelligence Laboratory at the University of Jyväskylä.

According to Äyrämö, the use of synthetic data also accelerates authorization processes and thus, among other things, the testing of new ideas.

See also  Caring for the health of animals and the environment, the key to the well-being of people

The use of synthetic data can lead to better outcomes in patient care and medical method development

Data-driven AI methods can be used to help doctors make diagnoses. Although the technical potential of AI is enormous, the amount of medical data is often insufficient. This is a major challenge for the development of medically effective methods.

“By mixing real and synthetic x-ray images, we improved AI-based osteoarthritis classification systems,” says Fabi Prezja, the PhD researcher responsible for developing the artificial neural network.

In the future, synthetic data may lead to better outcomes in the development of medical methods and patient care, especially for medical conditions where real patient data is limited.

“Furthermore, the neural network is able to modify synthetic X-ray images according to expert specifications. This capability is very powerful and allows potential future use for medical educational applications and stress testing for other AI systems,” adds Prezja.

The research was carried out in collaboration with the Central Finland Healthcare District, whose director, a professor of surgery Juha Paloneva, considers AI-based diagnostic methods as a valuable way to transfer knowledge from an experienced doctor to the work of to support a younger. one. doctor.

The adjacent image shows a set of screenshots of an animation showing how a synthetic x-ray can be edited to an expert’s specifications.

“AI can be used to reveal, for example, hard-to-detect signs of early osteoarthritis. However, AI methods for osteoarthritis continue to improve, so the work continues,” says Paloneva.



Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Social Media

Most Popular

On Key

Related Posts