A widely used infusion pump can be remotely hijacked, say researchers

A hospital infusion pump widely used in hospitals and medical facilities has critical security flaws that allow it to be remotely hijacked and controlled, according to security researchers.

Researchers at healthcare security firm CyberMDX found two vulnerabilities in the Alaris Gateway Workstation, developed by medical device maker Becton Dickinson.

Infusion pumps are one of the most common bits of kit in a hospital. These devices control the dispensing of intravenous fluids and medications, like painkillers or insulin. They’re often hooked up to a central monitoring station so medical staff can check on multiple patients at the same time.

But the researchers found that an attacker could install malicious firmware on a pump’s onboard computer, which powers, monitors and controls the infusion pumps. The pumps run on Windows CE, commonly used in pocket PCs before smartphones.

In the worst case scenario, the researchers said it would be possible to adjust specific commands on the pump — including the infusion rate — on certain versions of the device by installing modified firmware.

The researchers said it was also possible to remotely brick the onboard computer, knocking the pump offline.

The bug was scored a rare maximum score of 10.0 on the industry standard common vulnerability scoring system, according to Homeland Security’s advisory. A second vulnerability, scored at a lesser 7.5 out of 10.0 could allow an attacker to gain access to the workstation’s monitoring and configuration interfaces through the web browser.

The researchers said creating an attack kit was “quite easy” and “worked consistently,” said Elad Luz, CyberMDX’s head of research, in an email to TechCrunch. But the attack chain is complex and requires multiple steps, access to the hospital network, knowledge of the workstation’s IP address, and the capability to write custom malicious code.

In other words, there are far easier ways to kill a patient than exploiting these bugs.

CyberMDX disclosed the vulnerabilities to Becton Dickinson in November and to federal regulators.

Becton Dickinson said device owners should update to the latest firmware, which contains fixes for the vulnerabilities. Spokesperson Troy Kirkpatrick said the pump is not sold in the U.S., but would not say how many devices were vulnerable “for competitive reasons.”

“There are about 50 countries that have these devices,” said Kirkpatrick. He confirmed that eight countries that have more than 1,000 devices, three countries have more than 2,000 devices, but no country has more than 3,000 devices.

The flaws are another reminder that security issues can exist in any device — particularly life-saving equipment in the medical space.

Earlier this year, Homeland Security warned about a set of critical-rated vulnerabilities in Medtronic defibrillators. The government-issued alert said the device’s proprietary radio communications protocol did not require authentication, allowing a nearby attacker in certain circumstances to intercept and modify commands over-the-air.

Neurobehavioral health company Blackthorn pulls in $76 million from GV to treat mental disorders

There are numerous challenges to finding effective treatments for mental disorders. However, Blackthorn Therapeutics, a neurobehavioral health company using machine learning to create personalized medicine for mental health, is betting its technological approach to finding drugs that work will put it ahead of the competition. Lucky for them, GV and other biotech investors have shown they agree by adding another $76 million in Series B financing to the coffers.

Today, Blackthorn announced the close of its $76 million series B round from GV, Scripps Research, Johnson & Johnson Innovation and a bevy of other biotech investment firms, including Polaris Partners, Premier Partners, Vertex Ventures HC, Alexandria Venture Investments, Altitude Life Science Ventures, ARCH Venture Partners, and Biomatics Capita.

Blackthorn has been heads down the last couple of years on a clinical trial for a drug that could potentially treat mood disorders. In April, the company announced positive results from its phase I trial for the drug.

The company plans to use the funding to advance its clinical-stage programs for mood disorders as well as for potential treatment of autism spectrum disorder, advancing towards clinical investigation in 2020.

Brian Chee, a managing partner at Polaris Partners, Lori Hu, a managing director at Vertex Ventures HC, and Julie Sunderland, a managing director at Biomatics Capital have joined Blackthorn’s board as directors in conjunction with the funding.

Blackthorn also recently added two people to its executive team. Jane Tiller has joined as chief medical officer and Laura Hansen as vice president, corporate affairs.

“BlackThorn was founded to bring new therapies to patients by applying advances in computational sciences to address patient heterogeneity, one of the biggest historical challenges in the field of neuropsychiatric drug development,” said Blackthorn’s president and COO Bill Martin, Ph.D. “Three years later, insights from our data-driven approaches are yielding patient enrichment strategies that could increase probability of clinical trial success and improve patient outcomes. We are grateful for our investors’ support to continue advancing our platform and therapeutic pipeline as we build out a world-class team at the intersection of technology and clinical neuroscience.”

Creative Destruction Lab’s second Super Session is an intense two-day startup testbed

Canadian startup program Creative Destruction Lab (CDL) escapes succinct description in some ways – it’s an accelerator, to be sure, and an incubator. Startups show up and present to a combined audience of investors, mentors, industry players (some of whom, like former astronaut Chris Hadfield, verge on celebrity status) – but it’s not a demo day, per se, and presentations happen in focused rooms with key, vertically aligned audience members who can provide much more than just funding to the startups who participate.

North founder Stephen Lake on stage at CDL’s Super Session 2019.

Seven years into its existence, CDL really puts on a show for its cornerstone annual event (itself only two years old) clearly shows the extent to which the program has scaled. From an inaugural cohort of just 25 startups with a focus on science, CDL has grown to the point where it’s graduating 150 startups spanning cohorts across six cities associated with multiple academic institutions. It has consistently added new areas of focus, including a space track this year, for which Hadfield is a key mentor, as is Anousheh Ansari, the first female private space tourist to pay her own way to the International Space Station and the co-founder and CEO of Prodea Systems.

The ‘Super’ in Super Session

This is the second so-called ‘Super Session’ after the event’s debut in 2017. It includes roughly 850 attendees, made up of investors, mentors, industry sponsors and the graduating startups themselves. As CDL Fellow Chen Fong put it in his welcoming remarks, CDL’s Super Session is an opportune moment for networking, mentorship and demonstration of the companies the program has helped foster and grow.

A keynote track included talks by Ansari and Hadfield, as well as from Celmatix CEO and founder Piraye Beim, and a fireside chat with North founder and CEO Stephen Lake. Subjects ranged from the importance of the linkage between exploration and technology, to what Beim described as “probably the first CDL talk to include menstrual health, vibrators, incontinence, and menopause, all in the span of 15 minutes.” Lake meanwhile discussed the future of seamless human-computer interfaces, and Ansari discussed her work founding the XPRIZE program and the impetus behind the current moment and interest in private space innovation.

Celmatix CEO and founder Piraye Beim speaking at the 2019 Creative Destruction Lab Super Session in Toronto.

The variety in the keynote speaker mix and topic selection is reflective of the eclectic and comprehensive nature of CDL’s modern program, which scouts globally for prospective startup participants. Its six hubs then enter into a matching process with startups signed on to take part, where each scores the other and that leads to placement.

How CDL works

CDL’s originating thesis is all about supplying the limiting resource in a startup ecosystem; the thing which the program’s organizers think is the missing ingredient that differentiates Silicon Valley from any other innovation hub in the world. Namely, CDL theorizes that this missing ingredient is what CDL Associate Director Kristjan Sigurdson calls “entrepreneurial judgement.”

Sigurdson explains that this basically boils down to the ability to know what are the most important things you need to do as an entrepreneur, and in what order. The missing piece, he says, isn’t ideas, funding availability or a lack of effort – instead it’s the kind of judgement that results from experience. CDL’s model, which emphasizes five sessions held periodically during which a panel of mentors helps startups set three clearly defined objectives they can accomplish within the next eight weeks.

After each of these sessions, some triage occurs – essentially CDL mentors gathered in closed door meetings and are asked if they’d work with any of the startups that presented during the session. If startups don’t receive sponsorship in these closed door meetings, then they’re not asked to participate in the next session, and effectively are out of the program. All told, the program graduates around 40-45 percent of the startups that enter the program, Sigurdson said.

Group session with small group mentoring on site at Creative Destruction Lab’s 2019 Super Session in Toronto.

CDL is also a bit out of the ordinary in that it takes no equity from the startups it works with – it’s fundamentally an academic program, started by the University of Toronto, and its designed to provide real-world business cases for the school’s MBA students to work on. But it’s become so much more – providing mentorship and guidance as described, and also connecting researchers who often enter into formal advisory roles with CDL companies.

Sigurdson also noted that CDL has actually seen “much higher investment levels” vs. the average for more traditional incubation or acceleration programs. “It’s a program that I think allows companies to raise money much more organically even though it’s an artificial program we created,” he said, referencing CDL’s own comparative research.

Lab-grown and forged in fire

True to its name, Creative Destruction Lab in practice feels like a generative cauldron of ideas, shared with peers and industry specialists for debate, discussion and reformation. Sessions are remarkable to witness – where else are you going to see brand new companies get direct feedback from astronauts and representatives of global space agencies, for instance.

Creative Destruction Lab opening keynote for its Super Session 2019 event.

The model is unique, but clearly effective, and able to scale – as evidenced from its growth to what it is today, from its starting point in 2012, when one founder described it as ‘7 people in a room.’ The room featuring presentations from space track companies alone featured around 50 people in attendance for instance – almost all of which were top-flight industry leaders and investors, including Hadfield, Ansari, CDL alumni Mina Mitry of Kepler Communications, and prominent Toronto angel investor Dan Debow. Startups presenting in the space track included Wyvern, a hyperspectral imaging company; Mission Control, a startup that wants to be the software layer for Moon rovers; and Atomos, which is building space tug for extra-atmospheric ‘last-mile’ transportation solutions.

It’s easy to see why this program results in solid investment pipeline, given the profile of the sponsors and mentors involved. And it’s another strong stake in the ground for the claim that Canada’s startup scene, with Toronto as its locus of gravity, is increasingly earning (and outperforming) its reputation as a global center of innovation.

Modern Fertility raises $15 million to sell its hormone tests — and gather more fertility data from its users

Modern Fertility is a San Francisco-based company that sells fertility tests directly to consumers, but increasingly, those customers will be educating the company, too. Indeed, the two-year-old startup now plans to develop a database of anonymized data about its largely younger demographic.

A fresh $15 million in funding led by Forerunner Ventures should help. Forerunner founder Kirsten Green, who takes a board seat as part of the round, is known for countless savvy bets on a wide number of consumer brands that have taken off with users, from Dollar Shave Club to Bonobos to Glossier. With Forerunner’s help, Modern Fertility may well become a breakout hit, too, though potential customers should also understand its limitations before they click the “buy” button.

First, let’s back up. We’d originally written about Modern Fertility last year, when it began selling a kit from its website that’s sent to women’s doorsteps and allows them to gauge their levels of eight different reproductive hormones by using a finger prick. More specifically, the startup sends off its customers’ panels to CLIA-certified labs, where the tests are conducted, and most prominently, those tests are looking at the women’s level of AMH, or anti-mullerian hormone.

Why that’s relevant: every egg inside a woman’s ovaries sits within a fluid-filled sac full of cells that support egg maturation and produce hormones, including AMH. A woman’s AMH levels can provide one clue about how many of these sacs — or follicles — she has. That in turn provides a clue as to how many eggs she can release, or her ovarian reserve.

The point, says Modern Fertility’s cofounder and CEO, Afton Vechery, is to enable women to learn more about their bodies without having to shell out $1,500 to gain access to a similar picture by turning to a reproductive endocrinologist, of which there are relatively few.  According to the Centers for Disease Control and Prevention, there are roughly 500 infertility clinics in the U.S., and 2,000 reproductive endocrinologists.

Mixed feelings in medical community . . .

It’s a compelling pitch, especially given that women are putting off children longer for a variety of reasons, including to secure their financial future. In 2017, for the first time, U.S. women in their early 30s eclipsed younger moms to become the group with the highest birth rate, according to CDC data.

But there is room for pushback. The reality is that AMH and other tests can be conducted elsewhere, including by competing startups, for roughly the same cost that Modern Fertility is charging its customers. (Its kits originally sold from its website for $199; today, they sell for $159.)

Fertility testing is also generally is covered by health insurance plans because fertility problems can be linked to or caused by other health problems like endometriosis. (Not covered, typically: actual infertility treatments.)

A far bigger concern to some doctors is the unnecessary alarm that AMH screening may create for women who haven’t been diagnosed with infertility and who are less than 35 years old.

As Zev Rosenwaks, director of the Center for Reproductive Medicine at Weill Cornell Medicine and NewYork-Presbyterian, told the New York Times a couple of years ago, “All it takes is one egg each cycle . . . AMH is not a marker of whether you can or cannot become pregnant.”

Esther Eisenberg, the program director of the Reproductive Medicine and Infertility Program at the National Institutes of Health, has also said that AMH doesn’t dictate a woman’s reproductive potential. In fact, the NIH funded research in 2017 that found a “non-statistical difference” between low and normal AMH levels in a time-to-pregnancy study of women who were between the 30 to 44 years and who did not have a history of infertility.

Asked about such findings, Vechery, who was most recently a former product manager at the genetic testing company 23andMe, has clearly heard such criticisms. She readily acknowledges that AMH is “not an indicator of your ability to get pregnant right now in this moment,” adding that “it has so many other helpful benefits in thinking about your reproductive health in a much broader sense.”

Vechery also notes the company’s team of PhDs. She points to a clinical study that was published in The Green Journal (the official publication from The American College of Obstetricians and Gynecologists). She also speaks of Modern Fertility’s medical advisory board, which includes dedicated five medical doctors, including reproductive endocrinologists Nataki Douglas, a former associate professor at Columbia University Medical Center, and Scott Nelson, a professor at the University of Glasgow.

All are important pieces to building Modern Fertility, but it’s nevertheless worth mentioning that the company employs just two full-time PhDs currently.

Further, the company’s medical advisory members, including Nelson, are paid consultants.

As for the study, which Modern Fertility sponsored, it doesn’t actually prove anything about the power of AMH testing, though it does underscore that AMH, along with the seven other hormones the company measures on behalf of its customers, can be tested just as effectively with “fingerstick sampling” as a traditional blood draw.

The educator turns the tables . . .

Those curious about Modern Fertility — often younger women eager to get a jump on any later reproductive issues they may face — may well decide that information about their hormone levels is enough to part with the cost of a kit, which includes a one-on-one phone consultation with a nurse.

Interestingly, when they do, they’ll increasingly be asked to opt-in to questions about their health, lifestyles, and more. They may be asked repeatedly, too, as the company recommends that customers re-take the test yearly to track their hormones over time. Indeed, because so many of Modern Fertility’s customers do not have fertility issues, the company hopes to aggregate as much pertinent information from them as possible in order to complement the vast amounts of research that has been conducted on infertility.

“The fertility space needs to catch up, and a huge part of what we’re focused on is moving fertility science forward,” says Vechery. “So much research is primarily done on these women who are having issues; Modern Fertility is interested in flipping that around.”

It’s a strange state of affairs, but we’ve talked with several customers of the company in the past, and one can imagine them supporting it however they can, thanks in part to the sense of community that Modern Fertility has also been fostering. Among other things, for example, the company hosts get-togethers for customers in San Francisco so they can share their thoughts, their fears, and, presumably, their results.

As for whether Modern Fertility is also interested in selling that anonymized data as has happened at genetic testing outfits like Ancestry and Vechery’s former employer, 23andMe, Vechery insists that it will not, that the information will instead be used to inform the company’s product development.

Fertility startups have generally been on a fundraising tear, and little wonder. According to one estimate, the  global fertility services market is expected to exceed $21 billion by 2020. In fact, while venture capital has poured into everything from period-tracking apps to sperm storage startups, Modern Fertility has its own direct competitors, excluding obstetricians. Among these is KindBody, a New York-based startup that raised $15 million two months ago, and three-year-old, Austin-based Everlywell, which has garnered $55 million from VCs so far.

Notably, Modern Fertility represents Forerunner’s first foray into the so-called femtech space. Asked about Green’s involvement, Vechery notes she was particularly “excited about the community,” which Phil Barnes of First Round Capital, has also cited as the reason he wrote Modern Fertility an early check.

Ultimately, though, says Vechery, “Our business model is information, and I think for Kirsten, seeing what that trusted brand could do in women’s health and the conversations it could spark” was what she found most compelling about the company.

We understand why. We also can’t help but wonder if those conversations will drive some women to see — unnecessarily — the very specialists that Modern Fertility wants to free them of visiting.

Modern Fertility has now raised $22 million to date. Among its other backers are Maveron and Union Square Ventures as investors.

Pictured above: Modern Fertility cofounders Afton Vechery and Carly Leahy. Vechery is CEO; Leahy is the company’s CCO, or chief commercial officer.

What top VCs look for in women’s fertility startups

A number of promising women’s health tech companies have popped up in the last few years, from fertility apps to ovulation bracelets — even Apple has jumped into the subject with the addition of period tracking built into the latest edition of the watch. But there hasn’t been much in the way of innovation in women’s sexual health for decades.

In-vitro fertilization (IVF) is now a 40-year-old invention and even the top pharmaceutical companies have spent a pittance on research and development. Subjects like polycystic ovarian syndrome, endometriosis and menopause have taken a backseat to other, more fatal concerns. Fertility is itself oftentimes a mysterious black box as well, though a full 10% of the female population in the United States has difficulty getting or staying pregnant.

That’s all starting to change as startups are now bringing in millions in venture capital to gather and treat women’s health. While it’s early days (no unicorns just yet) interest in the subject has been jumping steadily higher each year.

To shine a better light on the importance of tech’s role in spurring more innovation for women’s fertility, we asked five VCs passionate about the space for their investment strategies, including Sarah Cone (Social Impact Capital), Vanessa Larco (NEA), Anu Duggal (Female Founders Fund), Jess Lee (Sequoia) and Nancy Brown (Oak HC/FT).

Sarah Cone, Social Impact Capital

Sarah Cone, Social Impact Capital

We’re interested in companies that create large data sets in women’s health and fertility, enabling personalized medicine, clinical trial virtualization, better patient outcomes, and the application of modern AI/ML techniques to generate hypotheses that discover new targets and molecules.

Sequoia-backed Whole Biome wants to heal your gut with medical-grade probiotics

Whole Biome has pulled in $35 million in Series B financing from a list of investing titans, including Sequoia, Khosla, True Ventures, the Mayo Foundation and AME Ventues — just to name a few. The goal? To heal what ails you using microscopic bugs.

Medical science has caught on in the last few years about the importance of gut health using these bugs (also known as probiotics). Now startups are pitching in using venture money to come up with new and novel ideas.

“We’re at a unique point in time as the field of microbiome biology converges with enabling cutting-edge technologies and bioinformatics that will open up a whole new world of innovative health products,” said Colleen Cutcliffe, Whole Biome’s co-founder and chief executive officer.

Cutliffe, who hails from DNA sequencing company Pacific Biosciences, along with her partners Jim Bullard and John Eid, built a platform able to compute information from varying populations and compare microbiome sequencing to get a clear picture of what’s missing in a patient’s flora for overall health.

The next step is to use the raised funds to launch a product for the management of Type 2 Diabetes.

Many of the prescription diabetes medications out on the market today can come with a load of side effects like upset stomach, dizziness, rashes or inability to consume alcohol. However, Whole Biome says their product will not have any side effects.

Slated for release in early 2020, the startup has conducted double-blinded, placebo-controlled, randomized clinical trials for a product that releases special probiotics into your gut with the goal of reducing glucose spikes.

“Whole Biome is creating novel, disease-targeting microbiome interventions that have the potential to improve the course of many of the significant health issues facing people today,” said Sequoia partner Roelof Botha. “They have built an integrated approach and a multi-disciplinary team across research, development and commercialization to unlock complex microbiome biology and create products with both clinical efficacy and unparalleled safety.”

To date, Whole Biome has now raised $57 million in funding.

Groupon cofounder Eric Lefkofsky just raised another $200 million for his newest company, Tempus

When serial entrepreneur Eric Lefkofsky grows a company, he puts the pedal to the metal. When in 2011 his last company, the Chicago-based coupons site Groupon, raised $950 million from investors, it was the largest amount raised by a start-up, ever. It was just over three years old at the time, and it went public later that same year.

Lefkofsky seems to be stealing a page from the same playbook for his newest company Tempus. The Chicago-based genomic testing and data analysis company was founded a little more than three years ago, yet it has already hired nearly 700 employees and raised more than $500 million — including through a new $ 200 million round that values the company at $3.1 billion.

According to the Chicago Tribune, that new valuation makes it — as Groupon once was — one of Chicago’s most highly valued privately held companies.

So why all the fuss? As the Tribune explains it, Tempus has built a platform to collect, structure and analyze the clinical data that’s often unorganized in electronic medical record systems. The company also generates genomic data by sequencing patient DNA and other information in its lab.

The goal is to help doctors create customized treatments for each individual patient, Lefkofsky tells the paper.

So far, it has partnered with numerous cancer treatment centers that are apparently giving Tempus human data from which to learn. Tempus is also generating data “in vitro,” as is another company we featured recently called Insitro, a drug development startup founded by famed AI researcher Daphne Koller. With Insitro, it is working on a liver disease treatment owing to a tie-up with Gilead, which has amassed related human data over the years that Insitro can use to learn from. As a complementary data source, Insitro, like Tempus, is trying to learn what the disease does in a “dish,” then determine if it can use what it observes using machine learning to predict what it sees in people.

Tempus hasn’t come up with any cancer-related cures yet, but Lefkofsky already says that Tempus wants to expand into diabetes and depression, too.

In the meantime, he tells Crain’s Chicago Business that Tempus is already generating “significant” revenue. “Our oldest partners, have, in most cases, now expanded to different subgroups (of cancer). What we’re doing is working.”

Investors in the latest round include Baillie Gifford; Revolution Growth; New Enterprise Associates; funds and accounts managed by T. Rowe Price; Novo Holdings; and the investment management company Franklin Templeton.

Famed founder Daphne Koller tells it straight: “With most drugs, we do not understand why they work”

Daphne Koller doesn’t mind hard work. She joined Stanford University’s computer science department in 1995, spending the next 18 years there in a full-time capacity before cofounding the online education giant Coursera, where she spent the following four years and remained co-chairman until last month. Koller then spent a little less than two years at Alphabet’s longevity lab, Calico, as its first chief computing officer.

It was there that Koller was reminded of her passion for applying machine learning to improve human health. She was also reminded of what she doesn’t like, which is wasted effort, something that the drug development industry — slow to understand the power of computational methods for analyzing biological data sets — as been plagued by for years.

In fairness, those computational methods have also gotten a whole lot better more recently. Little wonder that last year, Koller spied the opportunity to start another company, a drug development company called Insitro that has since raised $100 million in Series A funding, including from GV, Andreessen Horowitz and Bezos Expeditions, among others. As notably, the company recently partnered with Gilead Sciences to find medicines to treat a liver disease called nonalcoholic steatohepatitis (NASH) because of all the human data on the disease that Gilead has amassed over the years.

Later, Insitro may target even bigger epidemics, including perhaps Alzheimer’s disease or Type 2 diabetes. Certainly, it has reason to feel optimistic about what it can accomplish. As Koller told a group of rapt attendees at an event hosted by this editor a few days ago, “We’re now at a moment in history where a confluence of technologies emerged all at around the same time allow really large and interesting and disease-relevant data sets to be produced in biology. In parallel, we see  . . . machine learning technologies that are able to make sense of that data and come up with novel insights that can hopefully cure disease.”

It all sounds like talk we’ve heard before in recent years, but coming from Koller, one gets the sense that we’re finally getting close, despite the mysteries of human biology. Below are some excerpts from Koller’s interview with journalist Sarah McBride of Bloomberg. You can also watch their conversation below.

On why Insitro struck a partnership with Gilead (beyond that it could prove lucrative, with up to $1 billion in milestones attached to successfully developing targets for NASH):

There are fairly broad categories that our technology is well-suited for. We’re really interested in creating what you might call disease-in-a-dish models — places where diseases are complex, where we really haven’t had a good model system, where typical animal models that have been used [for years, including testing on mice] just aren’t very effective — and creating those ‘in vitro’ models to generate very large amounts of data that can be interpreted using machine learning.

There’s a whole slew of diseases that lend themselves to this type of approach. NASH was one of them, so partly it was the suitability of our technology to this disease, and partly it was that Gilead was just a really good partner for it because they have a whole bunch of human data from some of the clinical trials that have been running [which give us] access to two complementary data sources. One is what happens to the disease in large human cohorts, and one is what happens when you look at what the disease does in vitro, in the dish, then see if we can use what we see in the dish using machine learning to predict what we see in the human.

On how Insitro views data differently than big pharma companies:

Pharma companies say, ‘We have lots of data.’ And you say, ‘What kinds of data do you have?’ And it turns out they have dribs and drab of data, each stored on a separate spreadsheet in someone else’s laptop. There’s metadata that isn’t even recorded. For them, it’s like, ‘Yeah, I did the experiment and obviously I recorded what I had to because it doesn’t make sense to throw it away,’ but they don’t think of it as something you build a company on top of.

We come at it a completely different way. We say, ‘This is the problem that you’d like to solve. If only we had a model that could tell us the result of this experiment without having to do the experiment, because it’s costly or complicated or even impossible [because it would involve perturbing a living human’s gene].’  Well, machine learning has gotten really good at building predictive models if you give it the right data to train the model. So we’re in the business of actually building data for the sole purpose of training machine learning models. We think of [these models] like little crystal balls that would allow you to avoid doing [these more expensive or complicated] experiments.

On the impact of the National Institutes of Health’s “All of Us” research program, which is an effort to gather data from one million or more people living in the U.S. to accelerate research and improve health in part by logging individual differences in lifestyle, environment, and biology:

I would say if anything that the U.S. is a little late to the game on this one. There have been a number of national cohorts have already been generated in different countries; the two that are currently best developed are in Iceland and in the U.K, but there’s also one in Finland and one in Ireland and even in Estonia, where they’ve taken a large population from within that country and measured their genetics, but also measured a whole lot of properties about those people, including blood biomarkers and urine biomarkers and behavioral aspects and physical aspects and imaging. And so what you have now (in these countries) is a dataset that tells you, ‘Nature perturbed this gene,’ and, ‘We see this effect on the human.’

[In the UK, specifically, where they started their program five years ago and recruited 500,000 volunteers who agreed to physical and cognitive and blood pressure testing and images of the brain and the abdomen, among other things] it’s an incredibly rich data set [from which] discoveries are coming along on pretty much a weekly basis.

… This is valuable not just primarily for gene therapies but just as a way of identifying targets that actually make a difference, because most drugs that go into clinical trials fail. And by most, I mean 95 percent. And most drugs fail because they are targeting the wrong things. They are targeting proteins or genes that do not affect the disease they are supposed to affect. The recent, very visible failures of Alzheimer’s drug trials — actually several of them in a row — were almost certainly because the protein they were targeting, called amyloid beta, is just not the right causal factor in the disease.

On what researchers can do now with stem cells that would have been impossible even a few years ago:

[There are now] tools that have enabled the creation of not only large amounts of data but large amounts of biologically relevant data. So we used to do experiments on cancer cell lines . . . but it’s not a very disease relevant model. Today, we can take a small sample of skin cells and use what’s called the Yamanaka factor, to reprogram those cells to stem cell status, which are the cells that exist effectively in the womb. And those cells are capable of differentiating themselves into neural cells or liver cells or cardiac cells, and those are very disease relevant because they represent human biology; you can take those cells now from patients and from healthy people and see if there are differences in how they appear.

Readers, we could feature more of the transcript here, but we highly suggest watching the conversation with Koller. If you use this text as a leaping off point, you’ll want to start listening at around the 13-minute mark. It’s definitely worth the time to listen to what she has to say, including about cystic fibrosis, spinal muscular dystrophy in babies, and why the “mouse models” we’ve long relied on for a wide number of seemingly ubiquitous diseases “range from bad to really, really bad.” Hope you enjoy it.

Scientists pull speech directly from the brain

In a feat that could eventually unlock the possibility of speech for people with severe medical conditions, scientists have successfully recreated the speech of healthy subjects by tapping directly into their brains. The technology is a long, long way from practical application but the science is real and the promise is there.

Edward Chang, neurosurgeon at UC San Francisco and co-author of the paper published today in Nature, explained the impact of the team’s work in a press release: “For the first time, this study demonstrates that we can generate entire spoken sentences based on an individual’s brain activity. This is an exhilarating proof of principle that with technology that is already within reach, we should be able to build a device that is clinically viable in patients with speech loss.”

To be perfectly clear, this isn’t some magic machine that you sit in and its translates your thoughts into speech. It’s a complex and invasive process that decodes not exactly what the subject is thinking but what they were actually speaking.

Led by speech scientist Gopala Anumanchipalli, the experiment involved subjects who had already had large electrode arrays implanted in their brains for a different medical procedure. The researchers had these lucky people read out several hundred sentences aloud while closely recording the signals detected by the electrodes.

The electrode array in question

See, it happens that the researchers know a certain pattern of brain activity that comes after you think of and arrange words (in cortical areas like Wernicke’s and Broca’s) and before the final signals are sent from the motor cortex to your tongue and mouth muscles. There’s a sort of intermediate signal between those that Anumanchipalli and his co-author, grad student Josh Chartier, previously characterized, and which they thought may work for the purposes of reconstructing speech.

Analyzing the audio directly let the team determine which muscles and movements would be involved when (this is pretty established science), and from this they built a sort of virtual model of the person’s vocal system.

They then mapped the brain activity detected during the session to that virtual model using a machine learning system, essentially allowing a recording of a brain to control a recording of a mouth. It’s important to understand that this isn’t turning abstract thoughts into words — it’s understanding the brain’s concrete instructions to the muscles of the face, and determining from those which words those movements would be forming. It’s brain reading, but it isn’t mind reading.

The resulting synthetic speech, while not exactly crystal clear, is certainly intelligible. And set up correctly, it could be capable of outputting 150 words per minute from a person who may otherwise be incapable of speech.

“We still have a ways to go to perfectly mimic spoken language,” said Chartier. “Still, the levels of accuracy we produced here would be an amazing improvement in real-time communication compared to what’s currently available.”

For comparison, a person so afflicted, for instance with a degenerative muscular disease, often has to speak by spelling out words one letter at a time with their gaze. Picture 5-10 words per minute, with other methods for more disabled individuals going even slower. It’s a miracle in a way that they can communicate at all, but this time-consuming and less than natural method is a far cry from the speed and expressiveness of real speech.

If a person was able to use this method, they would be far closer to ordinary speech, though perhaps at the cost of perfect accuracy. But it’s not a magic bullet.

The problem with this method is that it requires a great deal of carefully collected data from what amounts to a healthy speech system, from brain to tip of the tongue. For many people it’s no longer possible to collect this data, and for others the invasive method of collection will make it impossible for a doctor to recommend. And conditions that have prevented a person from ever talking prevent this method from working as well.

The good news is that it’s a start, and there are plenty of conditions it would work for, theoretically. And collecting that critical brain and speech recording data could be done preemptively in cases where a stroke or degeneration is considered a risk.

Apple Watch ECG capabilities arrive for users across Europe and Hong Kong

Apple’s latest-generation Apple Watch doesn’t just have a curved display and a new industrial design, one of the major features of the Watch when it launched last year were its advanced health-tracking capabilities, particularly in regards to heart health and AFib detection.

Those features arrived in the US in December, but users abroad have had to wait. Today, Apple announced that the electro-cardiogram feature and irregular rhythm detection functionality is coming to 19 European countries and Hong Kong in the Watch’s latest update. These users will also gain access to the irregular rhythm detection features available on Watch models Series 1 and later.

Supported countries include Austria, Belgium, Denmark, Finland, France, Germany, Greece, Hong Kong, Hungary, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Romania, Spain, Sweden, the U.K and Switzerland.

Irregular rhythm detection is available on Apple Watch Series 1 and later. The ECG app is only available on the Apple Watch Series 4, but if you’re a European or Hong Kong-based user and curious of the new capabilities, update your Watch to 5.2 then open the Health app on your updated iPhone and go through the on-boarding process.