Nanoparticle researchers spend most of their time in one thing: count and measure nanoparticles. Every step of the road, they have to verify their results. They usually do so by analyzing microscopic images of hundreds of nanoparticles packed well. Counting and measuring them takes a long time, but this work is essential to complete the statistical analyzes required to perform the next synthesis of properly optimized nanoparticles.
Alexander Wittemann is a professor of Chemistry Coloid at the University of Konstanz. He and his team repeat this process every day. “When I worked on my doctoral thesis, we used a large particle counting machine for these measures. It was like a cash register and, at that time, I was very happy when I could measure three hundred nanoparticles per day,” Wittemann recalls, “recalls Wittemann,” , remember . However, reliable statistics require thousands of measurements for each sample. Today, the greatest use of computer technology means that the process can move much faster. At the same time, automated methods are very prone to errors, and many measurements must still be performed, or at least double verification, the researchers themselves.
A correct count: even with complex particles during the Coronavirus pandemic, the good fortune put Wittemann in contact with its doctoral student Gabriel Monteiro, who not only has knowledge of programming and AI, but also has connections with computer scientists. Wittemann and Monteiro developed a program based on the open -target source AI technology “segment of any model”. The program allows the nanoparticle counting backed by AI in a microscopic image and the subsequent automatic measurement of each individual particle.
“For clearly definable particles, the ‘Basin Method’ has worked quite well so far. However, our new method can also automatically count particles that have a weight of weights or caterpillar, which consists of strings of two or three superimposed spheres” , explains Wittemann. “This saves a lot of time,” he adds. “In the time that would generally need to complete a synthesis of particles and make the corresponding time measurements, we can now concentrate on particle synthesis and examine them under the microscope, while the AI system is responsible for most of the majority of rest.
In addition to this, the measurements of AI are not only more efficient, but also more reliable. The ai method recognizes the individual fragments with greater precision and measures them more accurately than other methods, even those who perform humans. As a result, subsequent experiments can be adapted and carried out with greater precision, which leads to the fastest success of the test series.
The research team has published the new AI routine, as well as the required codes and data of the open access study in Git-Hub and Kondata for other researchers to use and discuss them.
#accelerates #nanoparticles #research
