Scientists Just Discovered How Water Gets From Blood Into The Brain
It’s not what we thought.
18 JUN 2018
Scientists have identified the means by which water in our blood supply crosses over to the brain, where it becomes the cerebrospinal fluid that surrounds and protects one of our most precious organs.
Every day, about half a litre of water is transported from our blood in this way, but while researchers knew a thin tissue in the brain called the choroid plexus was involved in this process, they didn’t understand how so much cerebrospinal fluid (CSF) could be produced. Now, we may finally have the answer.
“It is brand-new knowledge on a very important physiological process involving by far the most complex organ in the human body,” says neuroscientist Nanna MacAulay from the University of Copenhagen in Denmark.
It used to be thought that osmosis and associated forces regulated how water passed into the brain to produce cerebrospinal fluid, but as MacAulay’s team explains, a number of existing studies have helped demonstrate that “osmotic water transport does not suffice to sustain the rates of CSF production consistently observed in mammals”.
To investigate what mechanism could be driving the phenomenon, MacAulay and fellow researchers studied a mouse model where the conditions required for osmotic water transport were absent.
By inhibiting osmotic water transporters in the brains of live mice, the team found a previously unknown ion transporter called the NKCC1 co-transporter; it turned out to be responsible for approximately half of CSF production.
If the same molecular channel can be manipulated in humans, it could provide a revolutionary way of accessing and controlling the water system of the brain, easing pressure without resorting to invasive surgical operations – such as drilling holes in the cranium to drain fluid, and even removing pieces of the skull.
“If we are able to target this ion and water transporter with medicine, it would affect a number of disorders involving increased intracranial pressure, including brain haemorrhage, blood clots in the brain, and hydrocephalus,” MacAulay says.
Of course, it’s always impossible to guarantee that the results seen in animal studies will be replicated in research with human patients, but here the team is hopeful, mainly because the protein structure of the choroid plexus is similar.
If they’re right, it could result in what the researchers describe as a “paradigm shift in the field”, giving us a therapeutic target to treat brain pathologies involving increased pressure in events such as strokes, along with other serious cases where patients currently have few (or zero) non-invasive options.
“Of course, it would be ground-breaking if we were able to use this mechanism as a target for medical treatment and turn down the in-flow of water to the brain to reduce intracranial pressure,” MacAulay says.
“At worst, the patient may suffer permanent damage and even die as a result of increased pressure. Therefore, this basic mechanism is an important find to us.”
Now that we know that CSF production depends on more than just osmotic gradients, the researchers say the next steps are to find how this water flow channel can be controlled in the basolateral membranes of cells, getting us another step closer to maybe one day helping patients whose brains are getting dangerously pressurised.
‘Revolutionary’ material charges 12-times faster than current lithium-ion batteries
Billions of pounds have been poured into research of the “revolutionary” material graphene, though few real-world applications have so far been realised.
A Catalan startup called Earthdas is aiming to address that by producing a graphene-based battery that it claims can charge 12-times faster than current lithium-ion batteries – potentially transforming the usability of electric vehicles by decreasing charging times from hours to just five minutes.
“Currently, cities are experiencing an obvious shift in terms of mobility,” Rafa Terradas, founder of Earhdas, told ZDNet.
“We’re all aware we must reduce the space occupied by combustion vehicles and incorporate innovative solutions to reduce pollution.
“Because graphene batteries charge so quickly, that opens the door to revolutionary solutions in terms of energy sharing between users.”
Beyond the faster charging times, Earthdas batteries also have a 60 per cent increase in capacity compared to current batteries.
Unlike other startups and research initiatives developing graphene-based batteries, Earthdas says it is the first to build batteries that are commercially viable and plans to produce the first 3,000 units in the second half of 2018.
Since its discovery in 2003 by researchers at the University of Manchester, graphene has also been touted to improve everything from water filtration to solar cells.
This is made possible by the remarkable properties of the one-atom thick material, which is 200-times stronger than steel, more conductive than copper and as flexible as rubber.
Such is its potential that the European Union’s biggest-ever research initiative is the Graphene Flagship, which commands a budget of around €1 billion (£800 million). The initiative claims graphene applications like Earthdas’ will generate “economic growth, new jobs and new opportunities” for Europeans.
Robotic cranes, genetic libraries and activated oil: How the cannabis industry is redefining hi-tech
Canada’s marijuana companies are becoming known for developing best practices for a nascent industry that, if they do it right, could leave them well-positioned to dominate around the globe
At first blush, the Canopy Growth Corp. operation in Smiths Falls, Ont., doesn’t look especially high tech. The building is an old industrial pile that used to be a Hershey chocolate factory and it was until recently at risk of being demolished.
But today, past the tastefully decorated lobby and the biosecurity barriers, the factory’s rooms are full of bright green marijuana plants under even brighter white lights, with little irrigation hoses snaking in between the pots.
It looks like what you probably think a marijuana grow-op should look like, but then Jordan Sinclair points out “the birdhouse” hanging from the ceiling, in a room where no bird should ever, ever be.
Sinclair, Canopy Growth’s communications director, calls it a birdhouse because that’s roughly what it looks like, but it is actually a sensor monitoring humidity and temperature.
Once you notice the first sensor, you’ll find them everywhere in the cannabis industry, doing vital work and producing reams of data as companies try to scale up for legalization later this year.
As that day approaches, Canada’s marijuana companies are becoming known for developing best practices for a nascent industry that, if they do it right, could leave them well-positioned to dominate around the globe.
A lot of the work almost seems common sense. Plenty of growing practices are borrowed from other forms of greenhouse agriculture, pharmaceutical production and plain old manufacturing.
Figuring out how it all fits together takes work, but the basics of cannabis production are fairly simple.
Large “mother” plants are grown to provide cuttings that turn into genetically identical baby clones. The clones are allowed to rapidly grow — like a weed — under those bright lights with plenty of water and nutrition.
Once the plants are big enough, the light cycles are changed to trick the plants into believing that summer is turning into fall. The cannabis plants flower in anticipation of winter, and those flowers contain high levels of tetrahydrocannabinol, better known as THC. The buds are harvested, the growing room is cleaned out and the cycle then repeats.
But at every stage of the process, Canopy Growth uses data to refine the process, such as tweaking light cycles or nutrients in the soil, to maximize potency and yield.
“Really, what we want to do is move the needle on how much poundage we can pull out of one room,” Sinclair said. “I remember really well the first time we cracked 100 pounds out of a room. And now, if we came under it, it would be a big internal review process of how that happened, because the norm is to pull well over 100 (pounds) out of every room.”
Companies such as Canopy Growth are also working hard on selective breeding to refine the genetics of their cannabis plants, further pushing up the yield and enhancing the desired end effects.
Industry players talk about companies that will set themselves apart in the future by having distinct “genetic libraries” of marijuana strains.
Canada’s marijuana companies are developing best practices for a nascent industry that could leave them well-positioned to dominate around the globe
There are more signs of innovation further down the production cycle. For example, there’s the lab where Canopy Growth is experimenting with new edible products, including some sort of marijuana-infused beverage.
“Can we do it so that one is carbonated and looks and feels like a beer?” Sinclair asks. “Can we do one that can compete with traditional wine categories?”
In another area, he asks that cameras be turned off so that he can show off the “coolest” thing: a production line that will put activated cannabis oil into softgel pills.
Cannabis is especially sticky, Sinclair points out, which is a nightmare for running it through manufacturing gear.
“We don’t want it on camera because we don’t want our competitors to see what we’re doing,” he said. “But these are processes that are being used in other industries to encapsulate things in gel caps. You know, a vitamin E capsule is not a game-changing technology, but it’s innovative, and it was really hard for us to apply all of that to cannabis.”
There’s no clear roadmap for how to do corporate-backed production, so the industry is being forced to make it up as it develops, which is giving rise to a generation of cannabis startups that support the big producers.
In Boulder, Colo., Front Range Biosciences Inc. is growing cannabis “tissue culture” — essentially making tiny marijuana clones out of tissue samples in a lab.
In Toronto and Vancouver, Resolve Digital Health Inc. is using software to try to refine the dosage for medical patients. And Ample Organics Inc., which just moved into larger office space in east Toronto, is selling a system of barcode printers and scanners, plus the software to stitch it all together, that tracks every plant and every gram of marijuana produced from “seed to sale.”
Ample Organics chief executive, Jon Prentice, boasts that 75 per cent of the licensed producers in Canada use the Ample Organics kit.
The kit is most useful in helping producers meet the rigorous inventory management regulations mandated by Health Canada, but a happy side effect is that it creates even more data for producers to analyze.
Prentice said he expects companies will start using data and machine learning to try to track customer trends, which, in turn, can drive production decisions.
But it’s not just cannabis that interests Ample Organics. The same tracking software could be used to improve food safety, for example, or help car companies track down customers in the event of a recall.
“We have the technology to do this everywhere,” Prentice said.
Others have also caught on to the notion that innovation in the cannabis industry could spin off into applications for other industries.
For example, Hound Labs Inc. in Oakland, Calif., said it has built a working breathalyzer that can detect THC in a person’s breath in the critical few hours after smoking cannabis.
Co-founder and chief executive Mike Lynn, who is also an emergency room physician, said Hound Labs is looking to test its device in the field with law enforcement partners later this year. He said it wasn’t easy to find a way to detect THC in breath.
“It’s literally in your breath, like, in parts per trillion, which is akin to putting 20 Olympic-sized swimming pools together and looking for one or two specific drops of water in all those pools,” Lynn said.
Aside from the obvious applications for impaired driving, Lynn said developing such a sensitive device opens up all kinds of other possibilities.
“If you have lung cancer, for example, your body is shedding all kinds of potential markers for that lung cancer,” he said. “Or maybe you’re taking drugs for other diseases and maybe there’s ways to monitor whether those drugs are effective for those diseases.”
A lot of other startups are also being driven to serve the cannabis industry, so much so that it can be hard to tell who’s peddling truly innovative new products, and who’s just trying to get in on the gold rush.
Sinclair, at Canopy Growth, said he expects to see the first cannabis bankruptcies later this year, and the biggest risk comes from supporting businesses.
“In any sector, if you’re in a startup mode, the most popular question you’re asking yourself is: Are people going to like this product? In cannabis, we already know, right? People like cannabis,” he said.
“On the ancillary side of the business, that question has not yet been proven out. If people are in a startup phase and they’re trying to prove a product that is a service or a tech or an add-on, they really have to figure out what the needs are for the people who are going to be buying that, and that’s going to be the big producers.”
Nevertheless, the result of all this activity has people in the industry talking about Canada as the launching pad for companies with a worldwide vision.
For example, Edmonton-based Aurora Cannabis Inc. is building big with global growth in mind. A phone interview with chief corporate officer Cam Battley gives a sense of the industry’s frantic pace: the only time he was available to chat was a Sunday afternoon and, over the course of a 45-minute interview, he paused at least a half-dozen times to fret about how many emails were coming in.
The company is building Aurora Sky, a high-tech marvel of marijuana production that sounds more like a pharmaceutical factory than an agricultural facility.
The building is designed to have a slightly higher air pressure than the normal atmosphere outside so that if there are any leaks, air flows out through the holes, thereby keeping contaminants from getting in. People never go into the production rooms where the plants actually grow.
“It’s all completely automated, so when it comes time to harvest, we have automated robotic cranes that go and pick up entire tables and take them out of the growing bay, into the harvest area,” Battley said.
We’re combining technologies to an extent that has never been done before
Cam Battley, Aurora’s chief corporate officer
The robotic arms in the facility have some nifty tricks up their sleeves, too. Their ultraviolet and near-infrared cameras scan the plants for any kind of stress, from pests to infection, that might prevent the plants from producing optimal flowers. The cameras can pick up on problems even before they are detectable by the human eye.
“We’re bringing in innovative technologies wherever we can, some of which we don’t speak about publicly because it’s a competitive advantage,” Battley said.
“We’re combining technologies to an extent that has never been done before. At Aurora Sky, once it’s fully operating, it’s going to be the most advanced and automated agricultural facility in the world.”
The Aurora Sky facility has a modular design and was built using the same type of materials that will go into the company’s planned million-square-foot Denmark facility — the idea being to position Aurora as a global leader.
Both Aurora and Canopy are already selling into Germany’s fledgling medical market, and they’re both counting on continued growth in the coming years.
“Most people are paying attention to consumer legalization (in Canada) because it’s such a big deal and it’s a major social change. And it is a big deal. It will take the market from 300,000 patients with a prescription to several million — probably five to seven million Canadians,” Battley said.
“But take a look at Germany, for example. That’s a country about two-and-a-half times our population, about 82 million people. And Italy, about 60 million people. They’re just at the outset of creating their medical systems.”
With their technological innovation, proprietary genetic strains and growing experience, Canadian companies are hoping that other countries will follow Canada’s lead by initially allowing the sale of medical marijuana and then perhaps entering full-blown legalization.
If that happens, these pioneers could have a competitive edge in global competition for years to come.
Alex Krizhevsky didn’t get into the AI business to change the course of history.
Krizhevsky, born in Ukraine but raised in Canada, was just looking to delay getting a coding job when he reached out to Geoff Hinton about doing a computer-science PhD program in AI at the University of Toronto. The fateful moment was when, as a graduate student, Krizhevsky and a fellow student named Ilya Sutskever, decided to enter the ImageNet competition, a test for AI consisting of a huge database of online images.
The competition, open to anyone in the world, was to evaluate algorithms designed for large-scale object detection and image classification. The point wasn’t just to crown a winner, but to test a hypothesis: with the right algorithm, the massive amount data in the ImageNet database could be the key to unlocking AI’s potential. The two grad students, working with Hinton as an advisor, decided to enter the 2012 competition using a fringe idea: an artificial neural network designed by Krizhevsky. The approach dominated the contest, beating every other research lab by a huge 10.8% margin.
They beat every other research lab by a huge 10.8% margin.Thus, the current AI boom was born. Google hired the three researchers to seed a new, major projects using neural nets; the technology’s decision-making prowess soon put the words “deep learning” on the lips of every founder and Silicon Valley executive. Other tech companies like Facebook, Amazon, and Microsoft started positioning their businesses around the tech.
Now Krizhevsky, following a four-and-a-half year stint at Google, is riding the wave he helped generate, by joining deep-learning startup Dessa as its technical adviser. Dessa, previously called Deeplearning, works with companies to overhaul their businesses with AI. For example, it worked with Scotiabank to develop a deep-learning system that identifies the signs of potentially-delinquent customers faster.
A highly non-obvious solution
Back in his grad school years, Krizhevsky was reading papers on an earlier algorithm invented by his advisor, Hinton, called the “restricted Boltzmann machine.” He had seen graphics processing units (GPUs) used with restricted Boltzmann machines, instead of central process units (CPU). He thought that if could use those GPUs on other kinds of neural networks with more layers (or, “deep neural networks”) he could ratchet up processing speeds of deep neural networks and create a better algorithm.The result was a neural network design to quickly beat other state-of-the-art benchmarks in algorithm accuracy.
Shortly after that discovery, in 2011, Sutskever, another of Hinton’s grad students, learned about the ImageNet dataset. It was more than a million images, specifically crafted for the kinds of computer-vision algorithms that the Toronto team were trying to tackle. “I realized that his code was capable of solving ImageNet,” says Sutskever. “A highly non-obvious realization at the time.”
Krizhevsky then used the enhanced capabilities of his GPU-sped code to train the neural network on the dataset. The higher calculation speeds allowed the network to process those millions of images in five or six days, rather than the weeks or even months it would have taken previously. All the extra data that could be processed enabled the neural network to have unprecedented sensitivity in telling the differences between objects in an image.
“Unlike many other researchers, he’s an engineer at heart.”Hinton was originally resistant to the idea, since the neural network still needed to be told which objects were in which images rather than learning the labels itself, but still contributed to the project in an advisory role. It took six months just to break even with what were then the image-classification benchmarks for ImageNet, and then another six to achieve the results the team submitted.
“[Krizhevsky] has an extremely deep understanding of [machine learning], and unlike many other researchers, he’s an engineer at heart,” says Sutskever, who is now director of research at OpenAI. “He has the ability to keep at a problem until it’s solved.”
Krizhevsky, who is soft-spoken and has never talked to the media before now, chuckles when recalling the weeks after the 2012 ImageNet results came out. “It became kind of surreal,” he says. “We started getting acquisition offers very quickly. Lots of emails.”
The “end goal of computer science”
The eventual neural-network framework was validated in a seminal research paper (pdf) in the field of AI, first presented at AI’s largest annual conference in 2012, after the ImageNet challenge. That study has now has been cited more than 24,000 times, according to Google Scholar.
The neural-network framework that resulted is now known colloquially as AlexNet, but it didn’t originally bear that name.
After the acquisition, the name was rightfully changed to AlexNet.After the ImageNet challenge, Google tasked an intern named Wojciech Zaremba—now head of robotics at OpenAI— with recreating Krizhevsky’s paper for the company. Since Google has a tradition of naming neural networks after their creators, the company’s approximation of Krizhevsky’s neural network was originally called WojNet. But then Google won the war for the rights to hire the researchers and acquire their technology. After the acquisition, the name was rightfully changed to AlexNet.
During his Google tenure, Krizhevsky worked on Google Photos and then became deeply entrenched in the company’s self-driving car project. In September 2017, he left the company—he lost interest in the work, he says.
At Dessa, Krizhevsky will advise and help research new deep-learning techniques. The company is looking to double in size to 80 employees in 2018. To Krizhevsky, it makes perfect sense that AI has become such a force in the tech world and beyond.
“Artificial intelligence is sort of the end goal of computer science,” Krizhevsky says. “Computer science is about automating stuff, and artificial intelligence is about automating everything.”
Giant lasers pass new milestone towards fusion energy
Physicists working at the National Ignition Facility (NIF) in the US say they have passed another important milestone in their quest for nuclear fusion energy. They have shown that the fusion energy generated by the laser implosion of a deuterium-tritium fuel capsule is twice that of the kinetic energy of the implosion. By further trebling the fusion energy, they say they will be close to the long-sought goal of an overall net energy gain.
The $3.5bn NIF trains 192 pulsed laser beams on to the inner surface of a centimetre-long hollow metal cylinder known as a hohlraum. Inside is a fuel capsule, which is a roughly 2 mm-diameter hollow sphere containing a thin deuterium-tritium layer. Each pulse lasts just a few nanoseconds and the lasers can deliver about 1.8 MJ of energy. This powerful blast causes the capsule to implode rapidly, creating immense temperatures and pressures inside a central “hot spot”, where fusion reactions occur.
The long-term goal is that the energy of neutrons given off by fusion can generate electricity. Before this is possible, NIF must show that it is possible to achieve ignition – the point at which fusion reactions generate at least as much energy delivered by the laser system. This involves self-sustaining reactions, in which the alpha particles that are also emitted during fusion give off enough heat to initiate further fusion.
After experiments done in 2009-2012 fell well short of ignition, Omar Hurricane and colleagues at NIF made significant changes to their strategy. They changed the shape of the laser pulses to create much more stable implosions. In 2014, these “high-foot” pulses each yielded up to 17 kJ of fusion energy (and later 26 kJ ) – exceeding the roughly 10 kJ created in earlier experiments.
Now, the team has modified the “high-foot” pulses and changed the composition of the outer layer of the capsules from plastic to carbon. The new material is three times as dense as the plastic, which means that laser pulses with a third of the duration can impart the same kinetic energy to implosions. Less helium gas is needed inside the hohlraum to prevent its walls from blowing in prematurely, which in turn makes for more stable implosions. And that means that more laser energy is ultimately converted into the kinetic energy of capsules’ collapse.
In 2017, the researchers obtained 54 kJ of fusion energy per laser pulse – as measured by the number of neutrons and alpha particles produced. This is twice the kinetic energy of the imploding capsules, which they established by measuring the implosion speed using X-ray radiography and by simulating the changing mass of the evaporating shell. In contrast, the 2014 experiments only just about recouped the kinetic energy.
Closer to the threshold
Team member Sebastien Le Pape says that the new experiments created a greater density and pressure within the hotspot and about twice as much heating by alpha particles. Although the latest energy output is less than a thirtieth of that needed for ignition, he points out that self-heating makes the fusion process highly nonlinear. What is crucial, he says, is generating a “burning plasma”, in which alpha particles dump more energy in the hot spot than is lost through radiation and electron conduction. Reaching this point, he estimates, will require a fusion energy of around 150 kJ. “We are much closer to that threshold than we were before,” he says.
The long road to ignition
The team is now using capsules and hohlraums with diameters about 10% larger than before. The larger capsules absorb more energy, which should make them collapse more quickly and generate more fusion reactions. Having carried out eight laser shots since January, he says the preliminary results look promising. “Nothing is telling us that we can’t make a burning plasma,” he says.
Le Pape believes a burning plasma could be achieved within two years if the group can solve additional engineering problems. As to how much longer it will then take to reach ignition, he refuses to speculate. “It is really hard to answer that question,” he says. “It depends on what challenges we find.”
Fusion experts outside NIF are enthusiastic but remain cautious. Steven Rose of Imperial College London says that the research is “a significant advance on previous work at NIF,” arguing that although it remains to be seen how much higher the fusion output can be pushed, the group’s step-by-step approach is “plainly the right one”.
The University of Oxford’s Steve Cowley says that the group is “beginning to understand better how to control the asymmetries that have plagued NIF,” but points that even if it does achieve ignition “many more steps” will still be needed to turn fusion into a practical source of energy.