Even strong microscopes can see more, process data faster, and process more data with the aid of artificial intelligence.
In a new study, published Friday in the journal Nature Methods, researchers used new machine learning algorithms to combine a pair of novel microscopy techniques.
The marriage dramatically accelerated image processing and yielded crisp, accurate results.
Researchers use light-field microscopy to image quick biological processes in 3D, such as the beating heart of a fish larva.
Since the procedure necessitates the compilation of vast volumes of images, image processing can take days. Frequently, the final outcome is devoid of resolution.
Light-sheet microscopy, another method, relies on a single 2D plane from a sample. The approach produces high-resolution images in much less time, but gathers less detailed data.
With the help of artificial intelligence, scientists were able to marry the two techniques.
“Ultimately, we were able to take ‘the best of both worlds’ in this approach,” Nils Wagner, one of the paper’s two lead authors, said in a press release.
“AI enabled us to combine different microscopy techniques, so that we could image as fast as light-field microscopy allows and get close to the image resolution of light-sheet microscopy,” said Wagner, a doctoral student at the Technical University of Munich Germany.
The new hybrid method utilizes light-field microscopy to capture detailed images of large 3D samples, while light-sheet microscopy helps train the machine learning algorithms to process the image data more efficiently.
“If you build algorithms that produce an image, you need to check that these algorithms are constructing the right image,” said co-author Anna Kreshuk, group leader at EMBL and an expert in biomedical image analysis.
The new algorithms used the high-resolution images captured through light-sheet microscopy to ensure they were accurately processing the light-field microscopy data.
“This makes our research stand out from what has been done in the past,” Kreshuk said.
Researchers suggest the architecture of their novel algorithms can be adapted for a variety of different types of microscope technologies.
Scientists said they expect their breakthrough to help biologists study a variety of important larval and embryonic processes.
“Our method will be really key for people who want to study how brains compute. Our method can image an entire brain of a fish larva, in real time,” said co-author Robert Prevedel, another EMBL group leader.