Lübeck - Researchers at the Fraunhofer Institute for Marine Biotechnology and Cell Technology are developing rapid test that should only require drop of blood to diagnose asthma. An artificial intelligence (AI) observes the way in which immune cells move in the drop.
Asthma should be diagnosed as early as possible in order to treat it adequately. However, this is often difficult, especially with smaller children. In the "Killasthma" project funded by the state of Schleswig-Holstein, the Fraunhofer scientists and industrial companies are working on this problem.
With new rapid test, diagnostics should only take 60 to 90 minutes. The basic idea behind the procedure is that the movements of the immune cells of asthmatics differ from those of healthy people.
"In asthma, the movement of immune cells is greatly slowed down when they experience an inflammatory stimulus," explained Daniel Rapoport, head of the Cell processing group at Fraunhofer EMB. The research group made use of this knowledge in the development of the test set.
The idea is to observe the immune cells from the drop of blood under specially developed holographic microscope for about 90 minutes and to use their movement patterns to assess whether asthma Disease is present. The microscope, also known as cell scanner, enables automatic, three-dimensional observation of the cells.
"We can observe 2,000 to 3,000 cells at the same time, which guarantees high statistical accuracy," says Rapoport. The movement patterns determined are then transferred to neural network. Self-learning algorithms analyze the blood cell movement patterns and calculate diagnostic index.
"Based on lot of training data, the neural networks learn patterns and can differentiate the profiles of sick and healthy people," says the Fraunhofer project manager. The procedure is not yet available for clinical routine. According to their own information, the Lübeck scientists are currently optimizing the hardware and training the AI.