[ad_1]
Nvidia and Scripps Research Translational Institute – a nonprofit research organization – are teaming up to develop artificial intelligence (AI) guided genomic processing and analysis tools. As part of the partnership announced today, Nvidia data scientists will meet alongside Scripps genomics experts and bioinformaticians to study end-to-end health care issues. .
"This is an extraordinary time in the field of healthcare and medicine, with the intersection of large datasets and our expertise in portable sensors and genomics," said Dr. Eric Topol, founder and director of Scripps Research Translational Institute and Professor at Scripps Research, in a telephone interview. "This is what the Institute was founded on 12 years ago."
The joint research of the companies will focus on complete genomic sequences (the complete sequence of the genome DNA of an organism), continuous physiological portable devices and other sensors, as well as on the prevention of diseases – in particular numeric detection prediction of atrial fibrillation, an irregular heartbeat that increases the risk of atrial fibrillation. stroke. Future work will focus on diseases and datasets not yet selected.
Scripps will provide the bulk of the research databases, one of which contains more than 1,000 continuous heart rhythm recordings (another includes all the genomic sequences of 1,400 people aged 80 and over who have not never been sick). And researchers from both companies will use a combination of tailored neural networks and pre-trained models in research data experiments. Assuming everything goes well, they will compile their work and tools and make them widely available through open source.
"The goal is to provide methods and infrastructure to the overall research," said Kimberly Powell, vice president of healthcare at Nvidia. "We are seeing quite a lot of progress [in the field]. Dozens of algorithms are being approved by the FDA, [and some of them] transform the workflow into radiology … Radiology and imaging have taken advantage of major advances induced by imaging and video. "
Powell is not wrong.
Companies such as Deep Genomics use machine learning to identify models of large sets of genetic data and establish links to cellular processes. And in 2017, Google released a free-source tool called DeepVariant, which uses machine learning to identify all the mutations that a person inherits from his parents. (He won first prize in a 2016 FDA competition to promote improved gene sequencing.) Meanwhile, mainstream genomics companies such as 23andMe have begun using machine learning to map the impact of genetic engineering. genetic material of individuals on phenotypes such as weight.
Pharmacogenomics – which examines the role of genetics in the context of a person's response to medication – is another area in which encouraging progress has been noted. Genoox, a start-up that raised $ 6 million in 2018 under the leadership of Triventures, an investor in the healthcare industry, leverages AI algorithms to analyze large amounts of genetic data, manages the points of data via a search engine and generates personalized and actionable recommendations.
"Artificial intelligence is extremely promising for transforming the future of medicine," Topol said. "With NVIDIA, we aim to establish a Center of Excellence in Artificial Intelligence in Genomics and Digital Sensors, with the ultimate goal of developing best practices, tools and an AI infrastructure for broader adoption and application by community of biomedical research. "
Today's announcement comes after Nvidia unveiled a partnership with Canon Medical Systems to promote the use of AI techniques in medical and related research. Earlier this month, King's College London launched a separate project to "accelerate discovery" of critical data strategies and accelerate deployment in clinics.
Source link
