Revolution in Braunschweig: AI recognizes molecules with 99 percent precision!

Revolution in Braunschweig: AI recognizes molecules with 99 percent precision!

Researchers at the TU Braunschweig have made remarkable advances in microscopy by leveraging artificial intelligence (AI) to revolutionize image analysis. Led by Professor Timo de Wolff and Professor Uta Schlickum, an interdisciplinary team has developed an AI capable of characterizing scanning tunneling microscope images with an impressive 99 percent accuracy.

The use of AI in microscopy not only opens up new avenues for research, but also significantly improves efficiency. Before the introduction of this technology, scientists had to manually search and count molecules on scanning probe microscopy images, which was very time-consuming. Now AI enables precise evaluation in seconds and could significantly speed up research.

The challenge of training data

Previously, the AI ​​required thousands of labeled training images to deliver reliable results. One of the challenges was to generate enough data for training with a small amount of original images. To solve this problem, the researchers created a computer-generated training data set that is hardly inferior in quality to the real images. This innovative approach has enabled AI to also recognize the chiral alignment of molecules on real data with 99 percent reliability.

Another notable aspect of the research is the increase in efficiency of catalysis and sensor technologies, which is made possible by the chirality-dependent analysis of the molecules. The photon scanning electron microscope in LENA used in this research provides atom-precise images and is a crucial tool for these types of studies.

A network for artificial intelligence

The project KI4ALL, in which several universities such as the TU Clausthal and the Ostfalia University of Applied Sciences are involved, aims to create a comprehensive network in the field of artificial intelligence. A central component of the project is the establishment of an AI hub that promotes exchange and networking. For this purpose, digital teaching content is created on current AI application and research areas, which are available as Open Educational Resources (OER). This offer is aimed not only at students, but also at anyone interested in expanding their skills in the field of AI.

The project is funded by the Federal Ministry of Education and Research and the state of Lower Saxony and is part of the “Artificial Intelligence in Higher Education” funding initiative. This shows the growing importance of AI in the academic and practical world.

Synergies between quantum mechanics and AI

The influence of quantum mechanics on developments in the field of artificial intelligence is another exciting aspect The knowledge is discussed. Quantum mechanics offers insights into fundamental laws of matter and energy that have the potential to fundamentally revolutionize data processing and machine learning. Synergies from these two dynamic research fields open up new horizons, for example through the application of quantum principles such as entanglement and superposition in AI systems.

These principles could greatly increase the efficiency of data analysis and find application in diverse areas such as climate modeling, pharmaceutical research and materials science. However, the combination of quantum computing and AI presents not only opportunities but also challenges, including the high computational requirements and the complexity of developing suitable algorithms.

Advancing research in these areas not only promises innovative approaches to solving complex problems, but also raises ethical questions about data security and the responsible use of information. The overlap between quantum mechanics and AI could expand the boundaries of current technology and open up new opportunities for future research directions.