CHENNAI: Five fetal brains, aged between 14 and 24 weeks, were cut into 20-micron slices – less than half the thickness of human hair – by scientists at IIT Madras to create a detailed three-dimensional atlas of over 5,000 images using indigenous technology at nearly a tenth of the cost of research in Western nations. The atlas, scientists said, will take scientists and doctors several steps closer to understanding the human body’s most complex organ that controls thoughts and emotions.
“No one has ever seen the brain this close,” said Professor Mohanasankar Sivaprakasam, head of IIT-M‘s Sudha Gopalakrishnan Brain Centre, who created these high-resolution images. The objective was to create a detailed map of the human brain across ages to understand the structure and functionality better. “This will open pathways for the diagnosis and treatment of various diseases such as Parkinson’s, Alzheimer’s, and stroke. So, we decided that the atlas must remain open source,” he said.
The centre acquired over 200 brains, both normal and those with diseases, and processed at least 70 of them into cellular-resolution digital volumes through centre’s high-throughput imaging platform for the dataset, Dharani. Researchers identified and marked over 500 brain regions, and these findings were accepted for publication in 132-year-old peer-reviewed Journal of Comparative Neurology. The idea to create high-resolution 3D images of brain came after talks in 2015 with alumnus and Infosys co-founder Kris Gopalakrishnan, who felt it would not just contribute towards alleviating diseases but also add insights into research on artificial intelligence and machine learning. “We must understand intelligence from a human perspective to create a better AI,” said Gopalakrishnan, who is one of the funders.
Journal of Comparative Neurology editor-in-chief Dr Suzana Herculano-Houzel said Dharani is the largest publicly accessible digital dataset of the human fetal brain, created with less than one-tenth of the initial funds that powered the Allen Brain Atlas. The cost of creating Allen Brain Atlas is $150-$200 million compared to $15 million for the Indian dataset.
