Creators of the popular protein-folding game, Foldit, are seeking help to design a treatment for COVID-19.
- Since being founded in 2008, the crowdsourced protein-folding game, Foldit, has helped solve many novel problems.
- In recent months, the Foldit team has presented its community with problems relating to COVID-19.
- Foldit founder, David Baker, says over 20,000 different designs for potential COVID-19 antiviral proteins have been submitted.
Foldit Lab Report 7: Quarantine Edition<span style="display:block;position:relative;padding-top:56.25%;" class="rm-shortcode" data-rm-shortcode-id="7d7e9081ebec48e1d95fd8ac4b7a53a1"><iframe type="lazy-iframe" data-runner-src="https://www.youtube.com/embed/fbvvunp_wPY?rel=0" width="100%" height="auto" frameborder="0" scrolling="no" style="position:absolute;top:0;left:0;width:100%;height:100%;"></iframe></span><p>While most Americans are self-isolating, which certainly helps stop the spread of COVID-19, Baker is asking Foldit gamers to help hunt for proteins that could stop the virus in its tracks. They're specifically seeking proteins that block the viruses's entry into new cells upon entering the human body. If successful, new antiviral drugs could be developed that would reduce the symptoms once you're infected. </p><p>Brian Koepnick, who works in Baker's lab and helps run Foldit, <a href="https://www.hhmi.org/news/citizen-scientists-are-helping-researchers-design-new-drugs-to-combat-covid-19" target="_blank">says</a> the diversity of responses they receive to problems posed is a necessary step in discovering new solutions. </p><p style="margin-left: 20px;">"We find that the creativity of crowdsourcing is really, really useful—if you ask 100 people to do something, they'll do it in 100 different ways. That's really valuable for us in protein design problems."</p><p>As COVID-19 plagues the entire planet, driving fear and uncertainty in citizens, at least there is precedent for this disease. We know that this type of virus infects cells through its spike protein, which latches onto certain cells and proliferate. Baker says that a protein that "grabs the coronavirus's spike protein might be able to run interference," preventing it from attaching to other cells and spreading. </p><p>Every puzzle Baker's lab publishes is online for roughly a week. They work with up-to-the-minute information about COVID-19; thus, the team is constantly updating its puzzles. According to Baker, a few entries seem promising—there have been 20,000 different designs submitted already—though as with any treatment, each design will require real-world testing. </p><p><span style="background-color: initial;">Baker notes that they've successfully crowdsourced strategies for dealing with flu, which brings hope that a treatment could be found in this situation. </span><span style="background-color: initial;">"In general, the coronaviruses appear to mutate less than influenza viruses. So that makes them a little bit easier of a target."</span></p>
Foldit players have come up with more than 20,000 different designs for potential COVID-19 antiviral proteins. Scientists plan to test 99 of the most promising designs (shown here) in the lab.
Photo: Foldit<p>This is truly an unprecedented moment in history. While researchers have worked on pandemics across the planet before, there has never been such a sense of urgency. Our global response to this coronavirus is likely to set the stage for how we treat diseases of this magnitude in the future. And as science writer Ed Yong <a href="https://www.npr.org/sections/health-shots/2020/04/01/825179922/fighting-covid-19-is-like-whack-a-mole-says-writer-who-warned-of-pandemic" target="_blank">says</a>, there is reason for hope. </p><p style="margin-left: 20px;">"The first steps so far have actually been encouragingly quick. A vaccine candidate has already entered early safety trials after a record breakingly short time from actually identifying and sequencing the genome of this new virus."</p><p>There is a long road from trials to implementation, Yong says. We're 12 to 18 months away from a vaccine. Still, the rapidity of this process has been aided by the sheer number of researchers simultaneously working on the problem. </p><p>Give the number of players on Foldit's platform, it's not about expertise as much as, in Baker's words, persistence and ingenuity. Citizen science is one of the greatest benefits of the digital age. In many ways, platforms like Foldit are leading the way to a new form of education. If you're interested in contributing, <a href="https://fold.it/" target="_blank">download the software and start playing</a>.</p><p>--</p><p><em>Stay in touch with Derek on <a href="http://www.twitter.com/derekberes" target="_blank">Twitter</a> and <a href="https://www.facebook.com/DerekBeresdotcom" target="_blank">Facebook</a>. His next book is</em> "Hero's Dose: The Case For Psychedelics in Ritual and Therapy."</p>
Medical researchers put a ring on it to learn more about the onset of COVID-19.
TemPredict<img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8yMjg5NjQxNi9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY0NDc2NzQ2NH0.IsTuW9DrQdLMVKpR5_oQmiT2b1J_oFbyjeSlDlkIIBs/img.jpg?width=980" id="930dd" class="rm-shortcode" data-rm-shortcode-id="98c3374bd834cc556e547c187893da35" data-rm-shortcode-name="rebelmouse-image" data-width="992" data-height="399" />
Image source: University of California at San Francisco Medical Center/ŌURA<p>The <a href="https://www.sealab.ucsf.edu/tempredict" target="_blank">TemPredict</a> study is a project of University of California at San Francisco Medical Center (UCSF) and Oura, in collaboration with Zuckerberg San Francisco General Hospital (ZSFGH).</p><p style="margin-left: 20px;"><em>The purpose of this study is to collect information from a wearable sensor that may allow researchers to develop an algorithm that can predict onset of symptoms such as fever, cough, and fatigue, which can characterize COVID-19</em>," reads a statement on the UCSF SEA Lab website.</p>
Image source: ŌURA<p>Data for TemPredict is sourced from a commercially available wearable, the <a href="https://ouraring.com" target="_blank">Oura ring</a>. The rings are typically marketed as activity monitors that help customers develop healthier sleeping habits. Nonetheless, they're packed with technology that the TemPredict team hopes can help them track the advance of COVID-19.</p><p>Each titanium Oura ring is equipped with infrared LEDs, NTC temperature sensors, an accelerometer, and a gyroscope to capture measurements such as heart rate, temperature, respiration, and steps. The rings are also accompanied by a smartphone app that collects the data that Oura and UCSF need for this project.</p><p>While no one claims that the Oura ring can detect COVID-19 in and of itself, that could change if the TemPredict team is able to successfully develop their diagnostic algorithm from the collected data.</p>
Getting into the project<p>TemPredict plans to equip emergency workers with an Oura ring if they don't already own and wear one. (If you're a UCSF or ZSFGH healthcare employee, there's a <a href="https://ucsf.co1.qualtrics.com/jfe/form/SV_3KtofTDPsT6BBmB" target="_blank">brief online questionnaire</a> that will tell you if you're eligible to join the study.)</p><p>Participants in TemPredict are expected to download the Oura app and connect it to their rings, which they agree to wear every day for three months after completing a screening and baseline survey.</p><p>On each day of the project, participants will fill out a survey recording any symptoms they've acquired, including fever, cough, fatigue, and other symptoms. In addition, they're expected to share the data their ring has collected — including temperature, heart rate, respiratory rate, sleep, and activity — with Oura, who will presumably forward it to the TemPredict data-crunchers. The company already collects data from some 150,000 of its rings worldwide and is also making that trove of data available to the TemPredict team.</p>
An elegant, 400-year-old means of navigating the stars takes flight.
- The Planetary Society is about to launch LightSail 2, a crowdfunded light sail craft.
- LightSail 2 uses photons from the sun as fuel.
- Space X's Falcom Heavy rocket will carry LightSail 2 aloft, 720 kilometers up.
Rising slowly<p><u></u>The pieces of Kepler's dream have been falling into place bit by bit since that letter to Galileo. The discovery of photons in the late 1800s by James Clerk Maxwell revealed the energetic particles in light whose momentum could be transferred to other objects. </p><p>Friedrich Zander envisioned the "tremendous mirrors of very thin sheets" propelling craft through space, and then Carl Wiley foresaw a solar sail as a shiny, reflective, parachute-like material opening in the direction of sunlight.<br></p><p>By 1976, Carl Sagan went on TV to show off a demonstration model of a light sail craft, <a href="http://www.planetary.org/multimedia/video/lightsail-then-and-now.html" target="_blank">enthusing</a> about the amazing technology and its potential. </p><p>Among Sagan's students some 40 years ago was Nye, a frequent Big Think contributor. The Society was founded by Sagan, Bruce Murray and Louis Friedman in 1980. In 2005, the Society launched the world's first light sail craft, the Cosmos 1, aboard a submarine-based ICBM. Unfortunately, it was lost when the ICBM failed before allowing Cosmos 1 a chance to fly on its own.</p><iframe width="560" height="315" src="https://www.youtube.com/embed/fCMQEUU4LHs" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
About the Planetary Society<p>The Planetary Society is the world's largest non-profit space organization, crowdfunded by over 50,000 members from over 100 countries, and supported by hundreds of volunteers. The Society was founded as outlet for the general public's interest in space, a level of interest not always reflected in governmental budgets. In addition to mounting <a href="http://www.planetary.org/explore/projects/" target="_blank">projects</a> such as the LightSail craft, the Society serves as an educational connection between the scientific community and the general public, advocates for robust governmental funding of space programs, and provides anyone an opportunity to get involved in some real space science.</p><iframe width="560" height="315" src="https://www.youtube.com/embed/1gny4Q9Kcl0" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
The Society’s Lightsail craft<p>At the center of each frankly beautiful LightSail craft is a cubesat. While we tend to think of satellites as large, bus-sized objects, they can be much smaller for simpler missions. The cubesat for the upcoming LightSail 2, for example, is about the size of a loaf of bread.</p><p>At launch, the cubesat and sails are encased in four solar panels. Once in orbit, the panels swing up into operational position, exposing the cubesat and stored sails.</p><p>The sails themselves are four shiny Mylar sheets 4.5 microns thick — that's thinner than a human hair. They're next pulled outward by four cobalt-alloy booms that extend like tape measures. The process takes about three minutes. When deployed, the triangular sails together form a square that's just 32 square meters, about the size of a boxing ring.</p><p>The primary force to be overcome by LightSail craft is <a href="https://en.wikipedia.org/wiki/Orbital_decay" target="_blank">atmospheric drag</a>, its collision with gas particles in the Earth's upper atmosphere. Think of it as friction that causes a satellite to slow and thus drop from orbit. In order for a craft to catch enough photon "propellant" — and to be high enough to get away from the upper atmosphere, its orbit needs to be above about 700 kilometers. </p><p>The Society has built two LightSail craft.</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTYwMTYyMy9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTYzNTc1NjgxNH0.nC9Si08mHK3gUxnhzrx5vCfQvqn54-NKDgJd5MrbZoE/img.jpg?width=980" id="f70fe" class="rm-shortcode" data-rm-shortcode-id="a14aa054fcb1627f154a0d9aa94923a1" data-rm-shortcode-name="rebelmouse-image" />
Image source: Planetary Society
LightSail 1<p>Around 2014, NASA offered the Society a free ride aboard an Atlas V rocket as part of the agency's Educational Launch of Nanosatellites (ELaNa) program. Even though the Lightsail craft would be placed into orbit below the necessary 700-kilometer height, the Society decided to use one of their LightSails to test the mechanics of the sail deployment system. </p><p>Dubbed "LightSail 1," the sails successfully unfurled, as this selfie taken by LightSail 1 attests.</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTYwNTAwMC9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY0MjI5Mzg1Nn0.wPhIRmAjQ10Jcs56O1-F1icM-3lO24MqUkl-L1Htk1Y/img.jpg?width=1245&coordinates=309%2C158%2C309%2C158&height=700" id="37af6" class="rm-shortcode" data-rm-shortcode-id="dbaf58e3c941d930d4bcbfc840a0215d" data-rm-shortcode-name="rebelmouse-image" />
Image source: Planetary Society
And now LightSail 2<p>The second craft, now known and "LightSail 2," was slightly modified — particularly its software — according to insights gleaned during the first mission. It's scheduled as of this writing to go up from Kennedy Space Center in Florida later this month aboard a SpaceX Falcon Heavy as part of the U.S. Air Force's STP-2 mission from Kennedy Space Center in Florida.</p><p>This time, LightSail 2 will be carried within another, slightly larger satellite, Prox-1, developed by students at Georgia Tech. The Prox-1 will be placed into orbit around 720 meters up, and a week later will launch LightSail 2.</p><p>After a few days of diagnostics, LightSail 2 will open up its solar arrays, and then a day later, unfurl its sails. Over the next month, it will continually re-position its sails relative to the sun to raise its orbit — this is the main part of the mission, the actual solar sailing. </p><p>Mission complete, the craft will orbit for about a year before drag takes its toll, and LightSail 2 burns up plummeting down through the atmosphere. During this year, its position will be tracked via <a href="http://www.planetary.org/blogs/jason-davis/2018/20180425-pew-pew-lightsail.html" target="_blank">ground-based laser ranging</a>, and it may be visible to the naked eye. The Society will offers an online dashboard that can tell you where and when to look up to se this most elegant spacecraft.</p><iframe width="560" height="315" src="https://www.youtube.com/embed/gUEuMQivNOo" frameborder="0" allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
The Human Diagnosis Project project is building the world's "open medical intelligence" system.
- The Human Diagnosis Project can develop medical diagnoses with startling accuracy.
- The platform combines the knowledge of medical professionals and artifical intelligence.
- The goal of the project is to provide open, readily available high-level guidance and training to health care professionals across the globe.
About Human Dx<p>The Human Dx project is a partnership between the social, public, and private sectors — in the U.S., it's a 501 (c)(3) not-for-profit/public-benefit corporation. According to Komarneni, Human Dx's business model is as free of cost to users as possible while still generating enough income to be self-sustaining. There are now nearly 20,000 medical professionals in almost 80 countries contributing. Among Human Dx's partners are, as the company states: the American Medical Association, the Association of American Medical Colleges, American Board of Medical Specialties, and the American Board of Internal Medicine. They're also working in collaboration with researchers at Harvard, Johns Hopkins., University of California San Francisco, Berkeley, and MIT.</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTMwNTU1NS9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY1OTg1MzY5OH0.ukAKUQTCScqw_fgqQHH69Rq5vQrbfXmduHV0sNR-LYw/img.jpg?width=980" id="37145" class="rm-shortcode" data-rm-shortcode-id="451df51295464068d0b7038d27c54b8d" data-rm-shortcode-name="rebelmouse-image" />
Open Intelligence<p>While diagnoses produced by Human Dx <em>do</em> bring together the opinions of multiple medical professionals, it's far from a simple voting system. It incorporates its own massive data set, machine learning, and artificial intelligence in addition to the input from medical professionals to develop its diagnoses. In designing their collective intelligence, says Komarneni, Human Dx had to first re-think the idea of open intelligence itself.</p><p>"We believe that open intelligence is the third form of open knowledge," he explains. The first was open source-protocols such as those on which the internet is based, as well as operating systems such as Linux. These protocols enabled the second form, open content: Wikipedia, data libraries, and so on. Open intelligence combines the first two: "And when you think about A.I. in the context of software," says Komarneni, "it really is code which is smartly delivering content to you based on what you put into the system."</p><p>The importance of open intelligence is that without it being available at low cost or free, the <a href="https://bigthink.com/robby-berman/how-artificial-intelligence-is-eating-democracy" target="_self" data-vivaldi-spatnav-clickable="1">cost of A.I. is going to be so prohibitive</a> that it'll "exacerbate, as opposed to close, income, health, and other disparities in society," warns Komarneni. Nowhere will the ramification be more serious than in health care, since "there is nothing we care more about than the well-being of the people we love and ourselves."</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTMwNzUyNy9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY2MzI0NDUwMH0.hCliFRrwCAOhXvnP0duOWPPC7_8D8Zm0OvwClMzpWGk/img.jpg?width=1245&coordinates=0%2C228%2C0%2C228&height=700" id="74a56" class="rm-shortcode" data-rm-shortcode-id="0306f87e39ac7352ddcb59691d16fb8f" data-rm-shortcode-name="rebelmouse-image" />
How Human Dx collective intelligence works<p>Collective intelligence in the Human Dx project is not unlike a panel of participants, when are referred to as "agents." Some of these are medical professionals, but they may also include the outputs of other systems. For example, Komarneni mentions that it's entirely possible IBM's Watson could be one of these agents, or even a data set from the National Institutes of Health.</p><p><strong>Lingua franca</strong></p><p>Of course, individual agents, even the human participants, express themselves in their own ways — is a lump "blue" or "blueberry-colored," for example — not to mention that contributions from some agents such as A.I. or datasets may be in the form of raw data. Before any meaningful synthesis of all these opinions can be performed, the first step is to convert them all into a common language of some sort. Human Dx's AI uses natural language processing, text prediction, and medical ontologies to derive these translations as the process's first step.</p><p><strong>Ranking opinions</strong></p><p>Human Dx establishes the capability, or CQ ("clinical quotient"), of each agent. To do this they rank agents' skills using test cases with known diagnoses, including "some of the most wickedly complex cases," says Komarneni. This allows Human Dx to determine how accurate agents' diagnoses can be expected to be, and how heavily they should be weighted against other participants' contributions in solving the current case.</p><p><strong>A.I. joins the panel</strong></p><p>At this point, the agents' inputs are synthesized to derive the most likely diagnosis, and this is combined in an A.I. model with all of the aggregated case data that's ever been captured by Human Dx — interactions in the "tens of millions" — including how "lots of other participants over many other cases have solved these cases." This A.I. model then joins the panel in arriving at the final diagnosis.</p><p>"And those [agents] combined," says Komarneni, "are how we can get to results that outperform the vast majority of individual participants."</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTMwNjg3Ni9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY2MDE3NDYxOX0.gBrdBnToROoodS9DwlpKnlmBSjl4EW6rkRPQNtiax9k/img.jpg?width=980" id="219e7" class="rm-shortcode" data-rm-shortcode-id="17eedce335a26c40615f098649d86b41" data-rm-shortcode-name="rebelmouse-image" />
The Harvard and Johns Hopkins studies<p>The Harvard study published in <em>JAMA</em> is the first public demonstration of the Human Dx system as a diagnostic tool. Working with an international cohort of medical students and professionals, the results were unquestionably amazing. There were 2069 users working 1572 cases — again, these were cases with known correct answers — from the Human Dx data set. About 60 percent of the participants were residents or fellows, 20 percent were attending physicians, and another 20 percent were medical students. In the study, as more medical professionals were added to the collective intelligence "panel," up to nine individuals, its accuracy consistently rose. Physicians who weren't specialists in their test-case areas achieved just a 62.5 percent accuracy score.</p><p>A previous study <a href="https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2720591?resultClick=3" target="_blank" data-vivaldi-spatnav-clickable="1">published</a> in <em>JAMA</em> in January, and done in cooperation with Johns Hopkins, looked at Human Dx as an automatic platform for assessing the diagnostic abilities of health care professionals and students. That the scores of participants looking at 11,023 case simulations were consistent with their training level shows, in Komarneni words, "that we provided a valid, quantitative, scalable measure of medical reasoning." While he admits this doesn't sound like a big deal, it is, since it offers a far more accurate and scalable option to current multiple-choice assessments, which have been shown to correspond poorly to real-world diagnostic skills.</p>
The future of health care and Human Dx<p>Komarneni says that there are basically only two ways to provide global universal health care, a pressing need since, "Almost half the world has no access to essential health services." One way, he says, would be to create a God-like A.I. system to provide health care to everyone, but, "We know that's not going to happen." God-like AI is just too hard, potentially requiring having to know everything about a patient from the tiniest details — say, the quantum behavior of electrons in mitochondria — to the huge, as in the kind of environment a patient lived in as a child.</p><p><span></span>In addition, Komarneni says, "In a world where data is locked up in many disparate silos, there isn't going to be a single collective agent. There's going to be a collective of many intelligent agents, both human and machine. The key is how do you integrate intelligence into larger buckets of intelligence than can solve the world's hardest problems."</p><p>This is where the Human Dx project, and the second approach, comes in. It actually has two components:</p> <ul> <li>The first is the expansion of existing medical professionals' diagnostic accuracy skills by providing them access to the Human Dx platform and its collective intelligence as a diagnostic tool.</li> <li>The second is helping to train new professionals, and <a href="https://www.Humanx.org/plans" target="_blank" data-vivaldi-spatnav-clickable="1">Human Dx Training</a> is already offering this on the Human Dx site.</li> </ul> <p>For those concerned with privacy in a system such as Human Dx, Komarneni says it'll be a non-issue, explaining with an example. When two people converse, "We don't have access to the underlying data of each others' minds. We're agents that are interacting with each other to gain relevant and useful information from each other." Similarly, Human Dx's system of interacting agents doesn't require the exposure of patients' personal data. What's shared with Human Dx are the conclusions agents draw from that data, not the data itself. In the case of a dataset operating as an agent, the data would be anonymized.</p><p>Human Dx's interest in all this is developing a platform it hopes others find uses for. "We believe we're just building the enabling technology that many other stakeholders could use." As examples, Komarneni imagines, "The VA could implement their own version of this. Kaiser Permanente could implement their own version. Employers could contract with us or with their own insurers. You could even also have individual and group practices use Human Dx software to serve patients directly."</p><p>Human Dx is currently looking at ways to open up as much of the project for non-professionals as possible, and they've already made a start: On their home page is a diagnosis cloud — mouse over the various blue bubbles to see different conditions, and then click for further details. In addition, just beneath the cloud is a search field with which you can look up diseases and symptoms.</p><img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTMwNTU4NS9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTY2NTE5NjI4NH0.xtH611B3ZlVU0NK0bRnspeUNohy2W6RZcFICZqu7jsw/img.jpg?width=980" id="d7e12" class="rm-shortcode" data-rm-shortcode-id="6eb97b433408586a3a24b90008891701" data-rm-shortcode-name="rebelmouse-image" />
Advances in satellite imagery are shining a light.
- Today, there are 40.3 million slaves on the planet, more than the number of people living in Canada.
- Slavery can be hard to find, but it commonly occurs in several key industries like fishing and mining.
- Using satellite data, researchers and activists are using crowdsourcing and artificial intelligence to identify sites where slavery is taking place.
Using an eye in the sky<p>Slavery takes place in the background, but its fingerprints are all over the products modern society relies on. Textiles, electronics, agriculture, and even brick-making all involve slavery to one degree or another. A growing body of research is using satellite imagery to shine a light on slavery practices. In fact, an estimated <a href="https://www.sciencemag.org/news/2019/02/researchers-spy-signs-slavery-space" target="_blank">one-third</a> of all slavery can be seen from space.</p><p>While rooting out specific instances of slavery can be difficult, we can use our knowledge of which industries include slave labor and pair it with satellite imagery and artificial intelligence to track down slavers and bring them to justice.</p><p>For example, AP's <a href="https://www.ap.org/explore/seafood-from-slaves/" target="_blank">Pulitzer-winning reporting</a> uncovered a vast slave network onboard fishing boats off the coast of Papua New Guinea. Although many of these ships were raided and hundreds of slaves were freed, other ships managed to escape. Its not too difficult to evade capture in the open ocean, but <a href="https://apnews.com/c2fe8406ff7145a8b484deae3f748aa5" target="_blank">DigitalGlobe</a> — a satellite company that provides Google Earth imagery — tracked down the rogue ships.</p><p>DigitalGlobe has also engaged in an effort to track slavery in fishing ships on Ghana's <a href="https://www.arcgis.com/apps/Cascade/index.html?appid=553180f8be4e40eb9e3caf8f28d25ba5" target="_blank">Lake Volta</a>. By inviting the public to pour through their satellite data, more than 80,000 ships, buildings, and fishing cages believed to be related to 35,000 enslaved children in the region have been tagged and mapped. As CEO Jeff Tarr stated, "You can't hide from space."</p><p>Satellite imagery has <a href="https://www.telegraph.co.uk/technology/2016/10/23/satellites-reveal-child-slave-camps-in-unesco-protected-park-in/" target="_blank">uncovered slave labor</a> in the Sundarbans mangrove forest in Bangladesh, where children clear the mangrove forests — critical to the ecosystem in that part of the world — as part of their forced labor processing fish. Still other work is being undertaken to observe mining sites that use slave labor, as well as numerous other industries where <a href="https://www.theguardian.com/global-development/2018/mar/19/experts-fight-slavery-satellite-pictures-south-asia-brick-belt" target="_blank">slavery is commonplace</a>.</p>
A new approach<img type="lazy-image" data-runner-src="https://assets.rebelmouse.io/eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpbWFnZSI6Imh0dHBzOi8vYXNzZXRzLnJibC5tcy8xOTIyMTEwNS9vcmlnaW4uanBnIiwiZXhwaXJlc19hdCI6MTYyMjYxODA1N30.3A0mOhvII-6kqAq7I943nwgGhiXCQ1xsiJz25ICa9k4/img.jpg?width=980" id="f6384" class="rm-shortcode" data-rm-shortcode-id="b664cb5c37f045b1dfb011d3fa75b6b3" data-rm-shortcode-name="rebelmouse-image" />
The region known as the "Brick Belt," where slave labor is frequently used, is outlined in red.
Boyd et al., 2018<p>While these attempts are all laudable, they represent just the beginning of a new satellite-based strategy to combat slavery. One of the biggest leap forward in the use of satellites to fight slavery is being undertaken by Doreen Boyd of the Rights Lab at Nottingham University. Her work focuses on the so-called "Brick Belt" that stretches crosses Pakistan, India, Nepal, and Bangladesh. This part of the world contains a large number of brick kilns. In the India region of the Brick Belt alone, an estimated 70 percent of brick kilns use slave labor.</p><p>In Boyd's <a href="https://www.sciencedirect.com/science/article/pii/S0924271618300479" target="_blank">previous work</a>, she used crowdsourcing and satellite data to gain an estimate of the number of brick kilns in the region. The number she reached was 55,387 kilns, a significant portion of which utilize slave labor, if expert estimates are to be believed.</p>
Results from an AI trained to identify brick kilns. The proposed brick kilns are surrounded by yellow boxes.
Foody et al., 2019<p>This is useful work: the problem of slavery can't be tackled in the region without identifying their locations, and one of these kilns has already been raided, resulting the freedom of 24 slaves. But more work is needed. Her previous study didn't identify the locations of all brick kilns, only a sample, and the region is too large to pore through manually. Crowdsourcing takes time and resources to complete and verify, and even if all brick kilns in the region were investigated, more would surely crop up in the future. Therefore, Boyd began to work on <a href="https://www.mdpi.com/2072-4292/11/3/266" target="_blank">developing an A.I.</a> that could identify brick kilns automatically from satellite data.</p><p>Machine-learning algorithms like the one Boyd used work by having humans "teach" the algorithm what it's looking for. Humans first tagged brick kilns from a small sample of satellite imagery; these are often circular or oval-shaped with a large chimney in the center. This sample was then fed to the machine-learning algorithm. Then, using the patterns identified by humans, the algorithm searched through other satellite data and pointed out places that matched the pattern. If the algorithm mistakenly selected areas that merely resemble brick kilns, those were used to refine the algorithm, teaching it what may a brick kiln is not.</p><p>In the small slice of the Brick Belt that Boyd and colleagues analyzed, their machine-learning algorithm identified 95.08 percent of the brick kilns in the region. While missing any potential sites of slavery — even just 5 percent — is a serious issue, the algorithm can be tweaked to overestimate the number of brick kilns. The advantage of this approach is that, although it would select many regions that were not brick kilns, it wouldn't miss any actual brick kilns either.</p>