This post is intended for my fellow machine learning engineers who are curious about applications in medicine, biology, or chemistry, but without a prior formal background in these fields. I have been in this position myself; my goal In this post is to give you a concise starting point into the field of drug discovery. That set the scene for the present post, where we are going to explore the "how" of actual approaches in greater detail. Unfortunately space does not permit a complete coverage of the rather large body of literature; I apologize in advance for subjectively leaving out some works and references. Nevertheless, I am going to discuss a number of prototypes that span the spectrum of approaches that have been developed so far.
A recent MIT Technology Review article titled "The Dark Secret at the Heart of AI" warned: "No one really knows how the most advanced algorithms do what they do. That could be a problem." Thanks to this uncertainty and lack of accountability, a report by the AI Now Institute recommended that public agencies responsible for criminal justice, health care, welfare and education shouldn't use such technology. Given these types of concerns, the unseeable space between where data goes in and answers come out is often referred to as a "black box" -- seemingly a reference to the hardy (and in fact orange, not black) data recorders mandated on aircraft and often examined after accidents. In the context of A.I., the term more broadly suggests an image of being in the "dark" about how the technology works: We put in and provide the data and models and architectures, and then computers provide us answers while continuing to learn on their own, in a way that's seemingly impossible -- and certainly too complicated -- for us to understand.
No list of artificial intelligence stocks is quite right without Alphabet (GOOGL, $1,169.44). Its Google division constantly pushes AI to new and exciting places. For example, a recent Bloomberg report has revealed that Google's Medical Brain unit is using AI to train machines to predict when patients will die. The Google tool uses self-learning neural networks to predict key outcomes including readmission and the length of hospital stay. This powerful data analysis can even be used to predict symptoms and disease, apparently with incredible accuracy.
Our brains are famously flexible, or "plastic," because neurons can do new things by forging new or stronger connections with other neurons. But if some connections strengthen, neuroscientists have reasoned, neurons must compensate lest they become overwhelmed with input. In a new study in Science, researchers at the Picower Institute for Learning and Memory at MIT demonstrate for the first time how this balance is struck: when one connection, called a synapse, strengthens, immediately neighboring synapses weaken based on the action of a crucial protein called Arc. Senior author Mriganka Sur said he was excited but not surprised that his team discovered a simple, fundamental rule at the core of such a complex system as the brain, where 100 billion neurons each have thousands of ever-changing synapses. He likens it to how a massive school of fish can suddenly change direction, en masse, so long as the lead fish turns and every other fish obeys the simple rule of following the fish right in front of it.
Smart wristbands, wireless sensing systems, and ultra-efficient solar cells – a glance through the list of winning projects from this year's ExploraVision competition might sound a lot like roll call at the Consumer Electronics Show. But, the concepts that claimed the top prizes aren't coming from tech's biggest names, or even the latest startups to break out of Silicon Valley. They're all, essentially, created by kids. 'Quite often young kids have these ideas for medical advancements, and it's out of empathy or feelings for someone they know – a friend of theirs, a family member who has some medical condition that they develop a device or system that would address it,' Nye said The Toshiba-backed initiative announced the winners of its annual k-12 science competition earlier this month, revealing the groundbreaking prototypes that seek to bring answers to our everyday problems. With solutions for everything from current medical failings to electrical grid woes, it's no wonder the student projects have already begun to capture industry attention.
After more than ten years of collaboration in a joint research partnership, Microsoft and Inria (National Institute for Research in Digital Sciences) have launched phase four of their initiative, with one key goal – to accelerate the deployment and adoption of Artificial Intelligence (AI) within France's technology ecosystem. This new phase of the partnership between Microsoft and Inria will focus on AI technology to unleash its potential, while putting it at the service of French companies. The launch focuses on two key framework components – research, and the transfer of skills and expertise. In addition to this research, both organizations will also share their knowledge and guidance with the startups of the AI Factory program, developed by Microsoft France, Inria, and Station F. Both partners will intensify the support received by these startups, enabling them to benefit from the expertise, learnings and applications resulting from the work carried out by researchers from Microsoft Research and Inria. The digital transformation projects deployed by Microsoft on behalf of major French companies will also accelerate their implementation and enable the French economy's flagship organizations to stimulate the development of new products and applications.
Why are we behind in neuroscience and the association with mental illness? Why are we behind in neuroscience and the association with mental illness? We're behind because mental illness is difficult to measure quantitatively, and the brain as a whole is both amazingly complex and poorly-understood. Like chronic pain, which is another tough-to-treat disorder, many symptoms of mental illness are subjective, and under these circumstances you need a very large number of patients and an enormous amount of rigor to do good clinical studies. The functional or structural correlates (i.e., the actual physical issues) underlying mental illness are only now coming to light with modern brain mapping technology, and often there is not a one-to-one correspondence between a certain structural change in the brain and a given diagnosis.
What if we could generate novel molecules to target any disease, overnight, ready for clinical trials? Imagine leveraging machine learning to accomplish with 50 people what the pharmaceutical industry can barely do with an army of 5,000. It's a multibillion-dollar opportunity that can help billions. The worldwide pharmaceutical market, one of the slowest monolithic industries to adapt, surpassed $1.1 trillion in 2016. In 2018, the top 10 pharmaceutical companies alone are projected to generate over $355 billion in revenue.
As she recovers from brain surgery, Simone Giertz is chronicling her experience with trademark honesty and humor. The "shitty robots" YouTuber has been tweeting steadily since her procedure in May, even posting a photo of her "super villain scar" a few days after surgery. On Thursday, she posted perhaps her realest update yet. SEE ALSO: 'Shitty Robots' creator Simone Giertz's TED Talk is a must-watch "Trying my best to make recovering from a brain tumor sound interesting but it's not," she wrote. "It's like trying to cook a meal out of a pile of sawdust.
Future Tense is a partnership of Slate, New America, and Arizona State University that examines emerging technologies, public policy, and society. This article is part of Update or Die, a series from Future Tense about how businesses and other organizations keep up with technological change--and the cost of falling behind. The patient was probably going to die. The one X-ray in town is often broken and isn't even at the hospital. Getting an accurate diagnosis is typically dependent on the clinical expertise of the health care professionals on staff and what they can ascertain with little more than a stethoscope.