Artificial intelligence techniques are becoming important tools in medicine and biomedical research. Tasks that require a great deal of precision, patience, and previous knowledge can now be taught to computers. A field where AI may be particularly useful is neurology, as the time scales and number of cells involved can be overwhelming for many experiments.
Appearing in the Proceedings of the National Academy of Sciences, the research involved speeding up two-photon calcium imaging, a well known method of recording neural activity. Currently, it is a laborious, mostly manual effort to identify which neurons are active during a stimulus. The researchers trained a system that, in their testing, was at least as good as humans at segmenting neurons, but it does so much faster than any human can.
Normally, a really good trained lab tech can segment a 30 minute video in about four hours. Others take longer. The AI system, on the other hand can do it in minutes, and surely can be made as fast as necessary thanks to the ubiquitous availability of cheap computing power.
“As a critical step towards complete mapping of brain activity, we were tasked with the formidable challenge of developing a fast automated algorithm that is as accurate as humans for segmenting a variety of active neurons imaged under different experimental settings,“ said Sina Farsiu, one of the researchers and an Associate Professor of Engineering in Duke BME.
“The data analysis bottleneck has existed in neuroscience for a long time — data analysts have spent hours and hours processing minutes of data, but this algorithm can process a 30-minute video in 20 to 30 minutes,” said Yiyang Gong, an assistant professor in Duke BME, who also worked on this research. “We were also able to generalize its performance, so it can operate equally well if we need to segment neurons from another layer of the brain with different neuron size or densities.”
“Our deep learning-based algorithm is fast, and is demonstrated to be as accurate as (if not better than) human experts in segmenting active and overlapping neurons from two-photon microscopy recordings,” said Somayyeh Soltanian-Zadeh, a PhD student in Duke BME and first author on the paper.
Study in PNAS: Fast and robust active neuron segmentation in two-photon calcium imaging using spatiotemporal deep learning…
Via: Duke…