A new deep learning method for biomedical Big Data (see below), achieved by scientists at the Gladstone Institutes and Google, and partly funded by the National Institute of Neurological Disorders and Stroke (NINDS), part of the National Institutes of Health, has been described as a breakthrough and “the tip of an iceberg that could transform biomedical science.” The Gladstone press release notes that, with the help of artificial intelligence, the number of features that can be obtained from images is nearly infinite. The limits of human imagination may be the only remaining factor.
“This is going to be transformative,” said Steven Finkbeiner, a director and senior investigator at the Gladstone Institutes. “Deep learning is going to fundamentally change the way we conduct biomedical science in the future, not only by accelerating discovery, but also by helping find treatments to address major unmet medical needs.”
“This approach has the potential to revolutionize biomedical research,” said NINDS program director Margaret Sutherland. “Researchers are now generating extraordinary amounts of data. For neuroscientists, this means that training machines to help analyze this information can help speed up our understanding of how the cells of the brain are put together and in applications related to drug development.”
Deep learning breakthrough for biomedical Big Data. Using Artificial Intelligence (AI) methods, scientists at the Gladstone Institutes and Google have developed a new approach to analyze massive amounts of high-resolution Big Data for biomedical research and applications. The scientists have used a deep learning algorithm dubbed “in-silico labeling” to analyze data, recognize patterns, and make predictions. The research work, published in Cell, shows that it may be possible to teach machines how to pick out features in neurons and other cells that have not been stained or undergone other damaging treatments. A deep neural network can predict fluorescence images from transmitted light images, generating labeled, useful, images without modifying cells. The study was partially funded by the National Institute of Neurological Disorders and Stroke (NINDS), part of the National Institutes of Health.
Artificial moles for early warning of cancer. Scientists at ETH Zurich have developed an early warning system for the four most common types of cancer. Should a tumor develop, a visible mole will appear on the skin. A research paper published in Science Translational Medicine indicates that, as soon as the calcium level exceeds a particular threshold over a longer period of time, which could indicate the presence of one of the four most common types of cancer (prostate, lung, colon and breast cancer), an implant inserted under the skin triggers the production of melanin, causing a mole to form.
Capturing detailed images and 3D movies of living cells. Researchers at Howard Hughes Medical Institute, led by Nobel laureate Eric Betzig, have found a way to merge lattice light sheet microscopy with adaptive optics, revealing the most detailed picture yet of subcellular dynamics in multicellular organisms. A study published in Science shows that a combination of the two imaging technologies permits watching in unprecedented 3D detail as cancer cells crawl, spinal nerve circuits wire up, and immune cells cruise through a zebrafish’s inner ear.
Remotely controlled gold nanorods activate immune T-cells in mice to fight cancer. Bioengineers at the Georgia Institute of Technology have installed a heat-sensitive switch into T-cells that can activate the T-cells when heat turns the switch on. The method, tested in mice and described in a study published in ACS Synthetic Biology, is locally targeted and could someday help turn immunotherapy into a precision instrument in the fight against cancer. The scientists have used gold nanorods, just dozens of atoms in size, to precisely warm cancer cells.
Advances in organoids for brain research. Scientists at Salk Institute have developed a new approach that can permit developing more sophisticated stem-cell-based brain-like “organoids” 3D models, used in brain research and drug development and testing, by ensuring the organoids receive sufficient oxygen and other nutrients via transplantation into rodents. The researchers are persuaded that the research work, published in Nature Biotechnology, could yield insights into the development of cures for brain disorders, speed up the testing of drugs, and even pave the way for someday transplanting healthy populations of human cells into people’s brains to replace damaged or dysfunctional tissue.
New option for implantable devices that inhibit pain. Using computer models and laboratory rats, Johns Hopkins researchers have demonstrated that direct electrical current can be delivered to nerves preferentially, blocking pain signals while leaving other sensations undisturbed. A study published in Science Advances describes a new concept for neural implants that use direct electrical current, long thought to be unsafe. According to the scientists, the experiments advance the search for improved implantable devices able to treat chronic pain that is due to peripheral nerve injury or disease.