[ad_1]
Artificial intelligence techniques are becoming important tools in medicine and biomedical research. Tasks that require a lot of precision, patience and prior knowledge can now be taught to computers. Neurology is an area in which AI can be particularly useful because the time scales and the number of cells involved can be overwhelming for many experiments.
Appearing in the Proceedings of the National Academy of Sciences, the research focused on accelerating two-photon calcium imaging, a well-known method of recording neural activity. Currently, it is a laborious effort, mainly manual, aimed at identifying active neurons during a stimulus. The researchers formed a system that, in their tests, was at least as good as human for segmenting neurons, but it is much faster than any human being.
Normally, a very good qualified lab technician can segment a 30-minute video in about four hours. Others take more time. The artificial intelligence system, on the other hand, can do it in minutes and can be achieved as quickly as needed thanks to the ubiquitous availability of cheap computing power.
"As a crucial step in the complete mapping of brain activity, we have been tasked with developing a fast automated algorithm, as accurate as humans, to segment a variety of neurons. badets imaged in different experimental contexts, "said Sina Farsiu. researchers and an engineering professor badociated with Duke BME.
"The bottleneck of data badysis has long existed in neuroscience – badysts have spent hours and hours processing data minutes, but this algorithm can handle a 30-minute video in 20 to 30 minutes, "said Yiyang Gong, badistant professor. Duke BME, who also worked on this research. "We were also able to generalize its performances. It can therefore work as well if it is necessary to segment the neurons of another layer of the brain with a different size or density of neurons. "
"Our deep learning algorithm is fast and has been shown to be as accurate as human experts in segmenting active and overlapping neurons from two-photon microscope recordings," said Somayyeh. Soltanian-Zadeh, a BME Duke and first author on paper.
Study in PNAS: Active and fast segmentation of neurons in two-photon calcium imaging by spatiotemporal depth learning …
Via: Duc …
[ad_2]
Source link
