<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Science + Communication + Policy - Medium]]></title>
        <description><![CDATA[Read about new research in medical devices, AI, and science policy - Medium]]></description>
        <link>https://medium.com/science-communication-policy?source=rss----c10e6d98d139---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Science + Communication + Policy - Medium</title>
            <link>https://medium.com/science-communication-policy?source=rss----c10e6d98d139---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 17:10:00 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/science-communication-policy" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Teaching AI to Talk to Clinicians]]></title>
            <link>https://medium.com/science-communication-policy/teaching-ai-to-talk-to-clinicians-88be5e26a35a?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/88be5e26a35a</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[healthcare]]></category>
            <category><![CDATA[medicine]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Thu, 25 Apr 2019 19:53:43 GMT</pubDate>
            <atom:updated>2019-04-25T19:53:27.472Z</atom:updated>
            <content:encoded><![CDATA[<p>When asked to explain how they arrived at a diagnosis, a good doctor can retrace their steps back to the moment the patient walked into the hospital. Sometimes they can go even further, back to the moment that brought the patient to the hospital in the first place. Diagnostics is a combination of test results, medical knowledge, and intuition (often referred to as clinical gestalt), and while your doctor may not always be right, the process of creating a diagnosis makes it possible to pinpoint the moment where the process went wrong.</p><p>However, defining this moment, and the diagnostic process in general, has presented a unique challenge for researchers developing machine learning for medicine. Developers rely on metrics like accuracy, sensitivity, and specificity to make sure that their models are performing correctly, only to encounter confusion from clinicians when they cannot explain how their model reached a particular conclusion. We saw this in 2018, when a report from cancer doctors working with an early version of IBM Watson Oncology’s software found themselves dissatisfied with Watson’s performance in the clinic and confused as to how it created cancer treatment plans.</p><p>Researchers at the University of Washington have begun to make strides in this area by developing <a href="https://www.nature.com/articles/s41551-018-0304-0">machine learning that explains itself to doctors</a>. In this case, their program, called Prescience, analyzes both patient history and real-time medical data from anesthetized patients during surgery to track a patient’s risk of developing hypoxaemia (low arterial blood oxygen tension), a condition that is linked to cardiac arrest, post-surgery infection, and stroke. Prescience then explains how it computed that risk value by highlighting clinical factors in a patient’s history (ex. BMI or age) and in the real-time surgery data (ex. pulse, blood pressure) that positively or negatively influence this calculation. This analysis is traditionally performed by an anesthesiologist based on their knowledge of the patient history and real-time data from standard surgery sensors, so Prescience is designed to act as an assistant in the operating room.</p><p>How can well Prescience identify signs of hypoxaemia? Well, it correctly predicted whether a patient would develop hypoxaemia about 81% of the time. If that sounds low, the researchers also studied how well anesthesiologists predicted whether a patient would develop hypoxaemia on the same medical data used to train Prescience, both by themselves and with the assistance of Prescience. Compared to Prescience alone, anesthesiologists without Prescience were correct about 66% of the time, going up to 77% when assisted by Prescience. In other words, Prescience was better at predicting hypoxaemia than anesthesiologists were, even when the anesthesiologists were assisted by Prescience in their decisions.</p><p>Does that mean that Prescience has found information in a patient’s history that doctors may overlook? Interestingly, it seems to rely on the same risk factors as doctors do — blood oxygenation, blood pressure, and whether the hospital specialized in trauma care were the three most important variables in the prediction of hypoxaemia onset for both anesthesiologists and Prescience. However, other variables such as BMI, pulse, and breathing rate were influential for Prescience but not for anesthesiologists, suggesting that Prescience may use correlations unseen by anesthesiologists when predicting the risk value.</p><p>While it is unlikely that you will see Prescience in your local operating room in the near future, you will likely see other applications of machine learning incorporated into your medical care if you haven’t already. One of the many hurdles that developers have faced and will continue to face in expanding the use of machine learning in healthcare is their ability to explain the decisions of a program to doctors and patients alike. Programs like Prescience demonstrate that the accuracy of prediction is only augmented by explanation, and that opening the black box of machine learning invites further discovery about programs and the people who use them.</p><p>Now, lean back in your chair, and count backwards from 10, 9, 8…</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=88be5e26a35a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/teaching-ai-to-talk-to-clinicians-88be5e26a35a">Teaching AI to Talk to Clinicians</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Machine Learning Bias Is Not a Partisan Issue]]></title>
            <link>https://medium.com/science-communication-policy/machine-learning-bias-is-not-a-partisan-issue-dc7a138647e7?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/dc7a138647e7</guid>
            <category><![CDATA[bias]]></category>
            <category><![CDATA[alexandria-ocasio-cortez]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ethics]]></category>
            <category><![CDATA[politics]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Thu, 24 Jan 2019 11:47:00 GMT</pubDate>
            <atom:updated>2019-01-24T11:50:52.465Z</atom:updated>
            <content:encoded><![CDATA[<p>On Tuesday, a video of Alexandria Ocasio-Cortez discussing bias in machine learning made its way around the Twitterverse. Specifically, she said that “[Algorithms] always have these racial inequities that get translated, because algorithms are still made by human beings.”</p><p>Ocasio-Cortez is one of the first politicians to highlight on the national stage the importance of the issues of fairness and bias in the development of machine learning algorithms. The challenge of AI governance continues to stump industry and academia alike. With no long-term solutions, a technology that would historically be regulated by the government is instead regulated by peer pressure. For those within the machine learning community, Ocasio-Cortez’s statements may have seemed like an opportunity to engage with lawmakers and shape legislation on an issue that presently has and will continue to have significant societal impacts.</p><p>Except her statements wasn’t framed as a legislative goal for a bipartisan Congress.</p><p>The trending video in which Ocasio-Cortez discusses machine learning bias was shared by Ryan Saavedra, a reporter at the Daily Wire. Saavedra’s past work has been critical of Ocasio-Cortez, with ledes including “Ocasio-Cortez Uses Violent Sexual Term To Describe Her Far-Left Agenda,” “Ocasio-Cortez Claims ‘GOP’ Thinks Her Dancing Was ‘Scandalous,’ Gets Roasted For Lying,” and, around the same time as his tweet, “AOC SNAPS: World Could End In 12 Years, Algorithms Are Racist, Hyper-Success Is Bad.<strong>”</strong></p><p>This critical perspective translated in his tweet, which framed the issue of combatting machine learning bias as <a href="https://twitter.com/RealSaavedra/status/1087627739861897216">socialist</a> in a tone that could be described as derisive. Concerns about framing machine learning bias as a “liberal” issue surfaced soon after.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C-hi0-3tmhMt3rQL0MD7RQ.png" /></figure><p>Let’s be clear: Combatting fairness and bias in machine learning is not Socialist. It’s not Democratic. It’s not Republican. It is political, because AI governance will likely require legislative action on the part of the federal government and/or governmental agencies.</p><p>In other words, it’s not a partisan issue.</p><p>We have been down this road before. Climate change used to be a bipartisan issue until the mid-1990s, when <a href="http://time.com/4874888/climate-change-politics-history/">industry lobbying succeeded in creating two sides on an issue where there is only one</a>. Scientists on the Environmental Protection Agency’s science advisory board were replaced with<a href="https://newrepublic.com/article/145582/scott-pruitt-declares-war-air-pollution-science"> industry advisors who downplayed links between air pollution and public health</a>. Under a Republican president and a Republican-led Congress, we’ve seen <a href="https://news.nationalgeographic.com/2017/03/how-trump-is-changing-science-environment/">extensive rollbacks</a> in environmental policy, including the U.S. withdrawal from the Paris Climate Agreement, which have handicapped our abilities to limit the effects of climate change on public health and the environment.</p><p>If allowed to continue being used on the public, bias in machine learning might be similarly catastrophic for at-risk populations. The evidence is already here — from <a href="https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing">biased predictions of recidivism rates </a>to <a href="https://www.wnycstudios.org/story/deep-problem-deep-learning">incorrectly classifying black women as gorillas</a> to <a href="https://www.theguardian.com/technology/2018/nov/12/deep-fakes-fake-news-truth">potentially malicious uses of deep fakes</a> to <a href="https://hub.packtpub.com/our-healthcare-data-is-not-private-anymore-study-reveals-that-machine-learning-can-be-used-to-re-identify-individuals-from-physical-activity-data/">re-identification of individual people from anonymized health data</a>. If biased machine learning is further integrated into our daily lives, marginalized communities will pay the price.</p><p>Some may argue that it is only one tweet, shared by one person. Except that’s not how social media always works. 24/7 hyper-connectivity has contributed to <a href="https://www.buzzfeednews.com/article/annehelenpetersen/millennials-burnout-generation-debt-work">generational burnout</a> and <a href="http://www.pewresearch.org/fact-tank/2017/10/05/takeaways-on-americans-growing-partisan-divide-over-political-values/">a deeply concerning growth in partisan divide</a>. It not easy to predict whether an idea on social media will fizzle out or become one of the defining issues of an era.</p><p>Right now, we need all sides — developers, ethics experts, theorists, scientists, doctors, historians, Republicans, Democrats, and everyone in between from all geographic and socio-economic groups — to come to the table and begin making decisions on AI governance. These decisions will be difficult, and will reveal the biases that make up every one of us. Adding partisanship into the equation will only ensure that more people suffer in the process.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dc7a138647e7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/machine-learning-bias-is-not-a-partisan-issue-dc7a138647e7">Machine Learning Bias Is Not a Partisan Issue</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Only Bendy in the Dark — Paper of the Week — May 30th]]></title>
            <link>https://medium.com/science-communication-policy/only-bendy-in-the-dark-paper-of-the-week-may-30th-d70b7dd4ae49?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/d70b7dd4ae49</guid>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[material-design]]></category>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[light]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sun, 30 Dec 2018 16:44:58 GMT</pubDate>
            <atom:updated>2018-12-30T19:42:36.219Z</atom:updated>
            <content:encoded><![CDATA[<h3>Only Bendy in the Dark — Paper of the Week — May 30th</h3><p>The paper can be found here: <a href="http://science.sciencemag.org/content/360/6390/772">Extraordinary plasticity of an inorganic semiconductor in darkness</a></p><p>*Disclaimer: I am far from a materials scientist. In fact, I’m probably closer to a layperson when it comes to inorganic materials and semiconductors. Comments and corrections are welcome. I’m always looking to learn!</p><p><strong>Welcome to Paper of the Week!</strong></p><p>Every week, I’m going to pick a paper that caught my eye (usually from Twitter if I’m being honest) and tell you about it!</p><p>This week’s paper is titled “Extraordinary plasticity of an inorganic semiconductor in darkness,” and it comes from a lab in the Department of Materials Physics, Nagoya University in Japan. This is definitely the first time I’ve ever seen the term “extraordinary” used in the title of a journal article, so this paper would have been a serious contender for that alone.</p><p>Let’s start from the basics. As you may have already noticed, some materials conduct electricity and some do not. Whether or not a particular material conducts electricity depends on the quantity of free electrons and holes (locations in the material with no electrons) in the material. Glass, which is an insulator, does not have free electrons, and therefore does not conduct electricity.</p><p>Thermal energy, or heat, can change the number of free electrons and holes present in a material, as can light irradiation. Increasing the number of free electrons and holes decreases the resistance of the material. If an electrical field is applied to this material (for example, by creating a voltage difference across the material), the holes and free electrons will become excited, and the material will become electrically conductive relative to the number of free electrons and holes. Semiconductors are these such materials, where some of the electrons are bound in covalent bonds, and some move freely within the material.</p><p>This is where darkness and plasticity come in. The number of free electrons and holes present in a material depends somewhat on the material’s crystalline structure, and that structure changes when stress or strain are induced into that material. Therefore, the properties of the crystalline structure of the material, including plasticity, can be altered with applied energy, in the form of thermal energy or light irradiation. In theory, you could have a material that becomes less conductive when bent, only for that reduction to be counteracted by light irradiation of the material.</p><p>But what about the absence of light? This paper found that a zinc sulfide crystal, when placed in total darkness, became significantly more plastic, or deformable. Crystals that were deformed under either LED or UV light reached roughly 2% strain before yielding, whereas crystals deformed in total darkness reached 45% strain. That’s a crazy amount of deformation for a crystal that could barely withstand stress without breaking in normal light!</p><p>We talked earlier about how light irradiation of a material could increase the number of free electrons and holes, so I would expect that while the total darkness might make the crystals more deformable, it would also reduce the energy of the free electrons and holes. Interestingly (to me, a person who does not work in this field), that’s not quite what happens. In the next section of the paper, which looks at the light-absorption characteristics of the material, the authors note that the optical band gap (or the amount of energy required for an electron to leave the covalent bond of the crystal structure and become free) decreases at 35% strain. While they did not show any data on conductivity tests between light-deformed and darkness-deformed crystal samples, I would think that this means that the conductance is similar between the two crystals, because of the reduction in the bandgap would allow previously free electrons to remain free. Alternatively, the conductance could decrease, but not as much as one would expect.</p><p>The paper goes on to discuss why this increased plasticity in darkness might occur, but in the interest of not misleading anyone, I won’t cover that since I didn’t fully understand it myself.</p><p>Why should you care? Well, semiconductors like zinc sulfide power many of the electronics we use today, and in recent years, we’ve seen technology begin to shift towards flexible electronics that could more easily interface with the average person’s body. If darkness allows for semiconductors to become more flexible while maintaining their conductive properties (or if that flexibility allows for the acquisition of new measurements due to changes in resistance based on deformation), then we could see some interesting new technology come of discoveries like this.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d70b7dd4ae49" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/only-bendy-in-the-dark-paper-of-the-week-may-30th-d70b7dd4ae49">Only Bendy in the Dark — Paper of the Week — May 30th</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Needs Citizen Science — Opinion]]></title>
            <link>https://medium.com/science-communication-policy/ai-needs-citizen-science-opinion-75d003f3cc79?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/75d003f3cc79</guid>
            <category><![CDATA[citizen-science]]></category>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[engagement]]></category>
            <category><![CDATA[science-education]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sun, 30 Dec 2018 16:44:37 GMT</pubDate>
            <atom:updated>2018-12-30T19:42:17.639Z</atom:updated>
            <content:encoded><![CDATA[<h3>AI Needs Citizen Science — Opinion</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*OvJuII5vwTghDNxur8kLLA.jpeg" /><figcaption>Me, as a freshman at Cornell, during one of my first few days in my undergrad lab.</figcaption></figure><p>Are you interested in doing science alongside academic and industry researchers?</p><p>No, you don’t need to have a background in science. All you need to participate is to enjoy playing games.</p><p>Chances are, you would be interested. In fact, <a href="http://www.pewinternet.org/2015/12/11/public-interest-in-science-health-and-other-topics/">roughly 60% of Americans expressed interest in science and technology in 2015</a>. For reference, 44% of Americans expressed interest in sports, so more Americans want to know about new and developing technology than want to know who won the last Superbowl (No, that’s probably not true, but you get the point).</p><p>Citizen science allows anyone to become involved in scientific research. In fact, it’s been around for a while, <a href="https://esajournals.onlinelibrary.wiley.com/doi/abs/10.1890/110278">dating back to the 17th century</a>, and has gained steam over the past several years as scientists learn to leverage the Internet and apps to reach wider audiences. Typically, citizen science allows the average person to help researchers collect and analyze data that they may not have access to otherwise. As exciting as data analysis likely sounds (Hint: It doesn’t), recent iterations of citizen science have focused on adapting data analysis techniques to online games. People have discovered protein structures by playing <a href="http://www.bakerlab.org/wp-content/uploads/2015/12/Khoury_Proteins_2014.pdf">FoldIt</a>, have identified new genes using <a href="https://phylo.cs.mcgill.ca/">Phylo</a>, and have sped up Alzheimer’s research on <a href="https://stallcatchers.com/main">Stall Catchers</a>.</p><p>The applications of citizen science are endless, especially in areas of research that rely on datasets that accurately represent groups of people, such as artificial intelligence.</p><p>Oh, wait.</p><p>Artificial intelligence doesn’t use citizen science?</p><p>Well, no, not exactly.</p><p>Currently, there are hundreds of different types of datasets available to AI researchers who are looking to train new algorithms. The data is often collected by research groups or industry companies and made public in an effort to support the wider AI community. However, recent headlines have detailed the failures and biases of AI that has been trained on those datasets: <a href="https://www.nytimes.com/2018/02/09/technology/facial-recognition-race-artificial-intelligence.html">from not recognizing images of black women</a> (or identifying black people as <a href="https://www.wired.com/story/when-it-comes-to-gorillas-google-photos-remains-blind/">gorillas</a>) to being more likely to classify <a href="https://www.propublica.org/article/how-we-analyzed-the-compas-recidivism-algorithm">black people</a> or <a href="https://rss.onlinelibrary.wiley.com/doi/abs/10.1111/j.1740-9713.2016.00960.x">low-income geographic regions</a>as at high risk for crime.</p><p>Some of these failures rely on access to data that the average person could not or should not provide, such as recidivism rates or HIPAA-protected health data. Access to that kind of data would require restructuring of data privacy regulations on a federal level, which should be advocated for both to provide better data to researchers and to protect the privacy of the data that is currently in use, but will not happen overnight.</p><p>However, there is no shortage of selfies on the phones and social media profiles of Americans, regardless of demographics.</p><p>To be sure, those participating in citizen science should be able to do so with the knowledge that their data will be protected and can only be used with their approval. <a href="https://www.forbes.com/sites/robertpearl/2018/04/10/blockchain-bitcoin-ehr/#3d725f7979e7">Blockchain</a> has been proposed as a potential solution to this, as has updated and more clearly defined privacy policies for citizen science research that handles personal data, such as the new <a href="https://www.cnbc.com/2018/03/30/gdpr-everything-you-need-to-know.html">GDPR</a> regulations.</p><p>Researchers need to engage citizens in the scientific process, both to improve technologies that will eventually be incorporated into their daily lives, and to maintain public support of science and technology during this time of increased public doubt in the scientific process. Create apps that allow people to see how their pictures have improved the accuracy of a facial recognition algorithm. Design games that identify biases in day-to-day decisions that researchers can account for in their research. Educate citizens on how they can become involved in science, and why they should, so that they can go back and tell their policymakers to support the research that might save and improve their lives.</p><p>After all, AI is the future, and the future is now. Don’t you want it to treat you well?</p><p>Photo: Frits Ahlefeldt from Flickr</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=75d003f3cc79" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/ai-needs-citizen-science-opinion-75d003f3cc79">AI Needs Citizen Science — Opinion</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[iBaby Monitoring — Paper of the Week — June 20th]]></title>
            <link>https://medium.com/science-communication-policy/ibaby-monitoring-paper-of-the-week-june-20th-b001f85fe8a5?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/b001f85fe8a5</guid>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[health]]></category>
            <category><![CDATA[neonatal-care]]></category>
            <category><![CDATA[medicine]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sat, 29 Dec 2018 16:31:50 GMT</pubDate>
            <atom:updated>2018-12-30T19:42:50.282Z</atom:updated>
            <content:encoded><![CDATA[<h3>iBaby Monitoring — Paper of the Week — June 20th</h3><p>The paper of the week can be found here: <a href="https://arxiv.org/pdf/1806.04037.pdf">Neonatal EEG Interpretation and Decision Support Framework for Mobile Platforms</a></p><p>Note: For the first time, this paper is open source! Special thanks to arxiv.org for existing and making it possible for the average person to stay up to date on scientific research.</p><p>Coming soon to a hospital near you — The iBaby Monitor!</p><p>This week, we’re talking about baby monitoring. Specifically, on your phone. This is probably not the baby monitor that you grew up with (unless you were born in the last couple months. If you were, you should probably alert an adult to the fact that you are an infant prodigy). It is also probably not one that you would use on your current/future children unless you want to find a nursery rhyme for when you have to connect EEG electrodes to their heads. However, it is a very simple solution to a problem that most people don’t realize exists that could become a real product without much effort, which is not something that we can usually say about recently published research.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*LDgVz9LXT_2oQyhfMyLWpw.png" /><figcaption>O’Sullivan et al. 2018</figcaption></figure><p>This paper, titled “Neonatal EEG Interpretation and Decision Support Framework for Mobile Platforms” came out of the Irish Centre for Fetal and Neonatal Translational Research (INFANT) (I see what you did there (: ) and the School of Engineering at the University College Cork. It details an Andriod app for mobile EEG monitoring of newborn babies in the Neonatal Intensive Care Unit (NICU). Sorry to all those iPhone users, although I’m sure this could be made into an iOS app too.</p><p>Why do babies need to be monitored? Well, if anyone saw the viral video of <a href="https://nypost.com/2018/06/19/child-scales-un-climbable-pool-ladder-in-viral-video/">a baby climbing up a closed and locked “unclimbable” pool ladder</a>, babies can clearly get into some trouble if left unattended. In this case, the authors are trying to help nurses in hospitals who have to take care of babies in the NICU, where premature babies or newborns with other medical issues live temporarily while they get treatment. One of the ways that nurses keep an eye on the babies is using electroencephalography, or EEG, which records the electrical activity of the brain. EEG can be used to identify and treat things like epilepsy, which can otherwise be hard to predict.</p><p>Now, nurses are pretty smart, but they also have a bunch of patients to care for every day, and they may not be trained to identify concerning trends in EEG data. Here’s where the Android app comes into play. The EEG is recorded using electrodes, and the data is sent to the Android app, which is on a tablet, through Bluetooth. The EEG data is filtered to remove noise in the signal (<a href="http://blog.catchpoint.com/2016/06/29/signal-vs-noise/">click here for another blog on signal noise and filtering</a>), and is displayed on the tablet. Interestingly, they use sound to help the nurses figure out whether the EEG signals are normal, with the app producing different sounds for different types of EEG signals. Finally, the app uses machine learning (<a href="http://cs231n.github.io/convolutional-networks/">specifically, a deep convolutional neural network</a>) trained on seizure and non-seizure EEG data to tell nurses whether the EEG data from a baby shows signs of an impending seizure.</p><p>So, why should we care? I mentioned this earlier, but in my experience, it is very uncommon to would make it easier for nurses to identify medical problems and monitor infants 24/7 without having to physically be there the whole time. Even better, it gives nurses an intuition for when EEG data looks normal and when something might be wrong. Best of all, it would not take a lot of work to give this kind of technology to hospitals if someone were to take this out of the lab. I’m looking forward to seeing (and hopefully doing) science that can easily move from the lab to the clinic in our near future.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b001f85fe8a5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/ibaby-monitoring-paper-of-the-week-june-20th-b001f85fe8a5">iBaby Monitoring — Paper of the Week — June 20th</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Biosensor Tattoos — Paper of the Week — July 5th]]></title>
            <link>https://medium.com/science-communication-policy/biosensor-tattoos-paper-of-the-week-july-5th-f13d479873dd?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/f13d479873dd</guid>
            <category><![CDATA[medicine]]></category>
            <category><![CDATA[biosensors]]></category>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[tattoo]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sat, 29 Dec 2018 16:30:39 GMT</pubDate>
            <atom:updated>2018-12-30T19:43:13.399Z</atom:updated>
            <content:encoded><![CDATA[<h3>Biosensor Tattoos — Paper of the Week — July 5th</h3><p>This week’s paper can be found here: <a href="https://scholar.harvard.edu/files/yetisen/files/the_dermal_abyss.pdf">The Dermal Abyss: Interfacing with the Skin by Tattooing Biosensors</a></p><p>I’m on Week 2 of Massive’s Storytelling 101 class, and this week was on taking the narrative style that we learned in the first week and using it on an actual paper. I chose a paper from the <a href="http://media.mit.edu/">MIT Media Lab</a>, which is always coming out with some odd or interesting research. The featured image on this post is a figure from the paper (all credit goes to the authors), so I’d definitely recommend checking it out. As always, feedback on my writing is always welcome!</p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p>Do you have a tattoo? One in five Americans do, using this approach to body modification to artistically represent an emotion, a moment, a person, or anything else. Sometimes tattoos can be imprinted in a way that makes it come to life with a movement or a particular pose. At the end of the day, however, a tattoo is a static image — it only changes as it fades from the skin.</p><p>Or does it? Researchers at the MIT Media Lab created tattoos that change color based on chemical reactions with the molecules in your body. Looking specifically at glucose and sodium levels, they injected solutions called biosensors into pigskin, then took pictures of the color changes when sugar or salt solutions were applied.</p><p>To get to this point, they started with a simple observation: Changes in skin color occur due to emotional reactions, certain diseases, and nutrition. Your face might get red when you’re angry, and yellow skin is one of the key signs that your liver is not functioning correctly. We can easily see these skin changes, and they help us identify that something is wrong before it gets worse. However, not all changes in your body result in skin color changes. Diabetics have to test their blood sugar regularly to catch any worrying trends. Runners and hikers have to supplement their hydration with water and electrolyte tabs to avoid dehydration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/666/1*EaNWA0o2PwOQseQXXADnZA.png" /><figcaption>Vega et al. 2017</figcaption></figure><p>Blood sugar changes and electrolyte changes do help us identify when our bodies need help, but to see those changes, we need invasive testing procedures. Researchers in the Media Lab, who had previously worked with molecules that fluoresce at different intensities with varying concentration of a specific compound, leveraged those fluorescent molecules to create tattoo “ink” that changes colors with changes in pH, glucose concentration, and sodium concentration. They then injected the inks into pig skin and added varying amounts of glucose, sodium, acid, or base, and watched the ink colors change in response.</p><p>While you probably won’t see this ink in your average tattoo parlor, this biosensor ink is the next generation of wearable technology for medical diagnostics and monitoring. Tattooed biosensors are a low-cost, easy-to-apply approach to monitoring patients, allowing those who cannot afford or those who otherwise do not have access to monitoring systems for their chronic medical problems to take control of their health. There is work to be done in making these biosensors longer lasting and more durable, but this is certainly a step in the right direction for new healthcare technology.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f13d479873dd" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/biosensor-tattoos-paper-of-the-week-july-5th-f13d479873dd">Biosensor Tattoos — Paper of the Week — July 5th</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hacking Healthcare — Why Our Future Medical Data May Be Under Attack]]></title>
            <link>https://medium.com/science-communication-policy/hacking-healthcare-why-our-future-medical-data-may-be-under-attack-a7443b9b80aa?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/a7443b9b80aa</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[medicine]]></category>
            <category><![CDATA[adversarial-network]]></category>
            <category><![CDATA[health]]></category>
            <category><![CDATA[science-communication]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sat, 29 Dec 2018 16:28:59 GMT</pubDate>
            <atom:updated>2018-12-30T19:43:46.535Z</atom:updated>
            <content:encoded><![CDATA[<h3>Hacking Healthcare — Why Our Future Medical Data May Be Under Attack</h3><p>This week’s paper can be found here: <a href="https://arxiv.org/pdf/1804.05296.pdf">Adversarial Attacks Against Medical Deep Learning Systems</a></p><p>Healthcare costs have been increasing to the point where medical care may become unaffordable for a significant portion of the US population in the near future. Many solutions have been proposed to slow down or reverse these cost increases, including passing federal healthcare programs (such as the ACA) that make medical care more efficient to limiting medical appointments to absolute emergencies.</p><p>One of these solutions is the idea of implementing artificial intelligence into medical practice. Ideally, this would reduce time spent on menial or recurring tasks, prevent inefficiency in the healthcare system, predict chronic conditions before they become irreversible, and provide better medical care to patients across the board. In fact, <a href="https://www.syracuse.com/opinion/index.ssf/2018/06/use_artificial_intelligence_to_rein_in_health_care_costs_commentary.html">I’ve advocated for this solution in an opinion piece in the past.</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dWBuA3_ya0ELecH-mMyp7Q.png" /><figcaption>Finlayson et al. 2018</figcaption></figure><p>However, researchers at MIT and Harvard Medical School have recently published work warning us that the healthcare system may not be ready for deep learning. As with any other program, medical deep learning systems can be attacked by those intending to steal private patient data, and their work reveals that medical systems may be more susceptible than most to a specific type of hacking — adversarial attacks.</p><p>The power of adversarial attacks has been seen in many areas of deep learning, including in a test of <a href="https://www.theverge.com/2017/11/2/16597276/google-ai-image-attacks-adversarial-turtle-rifle-3d-printed">Google’s deep learning-based image recognition software </a>(Spoiler: It failed). By making small changes to data that might not change a person’s classification of the data, you can fool deep learning algorithms into misclassifying information, resulting in incorrect patient diagnoses and potential patient harm.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a7443b9b80aa" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/hacking-healthcare-why-our-future-medical-data-may-be-under-attack-a7443b9b80aa">Hacking Healthcare — Why Our Future Medical Data May Be Under Attack</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Nanoscientists and Ethics — Paper of the Week — June 16th]]></title>
            <link>https://medium.com/science-communication-policy/nanoscientists-and-ethics-paper-of-the-week-june-16th-da2c1d980c8f?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/da2c1d980c8f</guid>
            <category><![CDATA[nanotechnology]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[ethics]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sat, 29 Dec 2018 16:26:51 GMT</pubDate>
            <atom:updated>2018-12-30T19:44:00.238Z</atom:updated>
            <content:encoded><![CDATA[<h3>Nanoscientists and Ethics — Paper of the Week — June 16th</h3><p>This week’s paper can be found here: <a href="https://asu.pure.elsevier.com/en/publications/scientists-ethical-obligations-and-social-responsibility-for-nano-2">Scientists’ Ethical Obligations and Social Responsibility for Nanotechnology Research</a></p><p>Note: Yes, I’m late this week. Normally, I try to have things out by Wednesday, but this week got a little crazy with planning for the Student Summit (if this is your first time on my blog, WELCOME and more details on the student summit can be found <a href="https://harrodjordan.github.io/posts/ScienceYouthSummit">here</a>), so I’m just now getting to all the other things. Planning to get back on track this week!</p><p>Also, this is another paper that is behind a paywall. In the future, I’m going to do my best to pick papers that are open sourced, but you don’t need the paper to get the idea for this one.</p><p><strong>Welcome back to Paper of the Week!</strong></p><p>For the first time ever, I did not find this paper on Twitter. In fact, it is not a new paper — it was published in 2016. It’s also not on “hard science” — instead, we’re talking about science advocacy.</p><p>Why am I writing about this? Well, in an era where new technology touches the average person’s life well before it is possible for governments to regulate it, scientists are one of the few groups who have the power to explain the possible consequences of using that technology, positive or not. However, communication and advocacy have not been a huge part of the academic community in the past, even between scientists, so we haven’t seen scientists step up to the plate as much as (I believe) they could. Similarly, science is often driven by whether or not we can do something, not whether or not we should, and I think that the latter should become a bigger part of the scientific method. Just because we can develop new technology, doesn’t mean that we shouldn’t be mindful of the ways that technology can be misused.</p><p>On to the paper: Today, we’re focusing on nanotechnology. This paper, titled “Scientists’ Ethical Obligations and Social Responsibility for Nanotechnology Research,” was written by Dr. Elizabeth Corley, Dr, Youngjae Kim, and Dr. Dietram Scheufele at Arizona State University and the University of Wisconsin-Madison.</p><p>Corley, Kim, and Scheufele note off the bat that policy change often lags behind emerging technology, allowing new science to go unchecked by the government, and that there have been some progress within the scientific community with regard to policy advocacy. Similarly, there have been studies on how the average person’s opinions on new technology depend on their personal values. However, there had not been much research into how scientists and engineers’ opinions on new technology are affected by their values. Luckily for us, Corley, Kim, and Scheufele decided to ask some nano-scientists, and report their findings.</p><p>What did they find out? Well, a lot of things:</p><ul><li>Female nano-scientists had a stronger sense of social and ethical responsibility for their research than male scientists.</li><li>Academic scientists were more likely to say that lab directors had an ethical obligation to keep their staff safe than non-academic scientists.</li><li>Academic scientists were more likely to say that scientists should be allowed to do research if they respected ethical standards than non-academic scientists</li><li>Non-academic scientists were more likely to say that federal funding agencies should require the labs the fund to have guidelines to protect their workers than academic scientists</li><li>Sense of social and ethical responsibility did not vary with political affiliation</li><li>Scientists who support nanotechnology regulations at the local or state level had a weaker sense of responsibility</li><li>Scientists who paid more attention to media coverage of social/ethical issues were more likely to think that authorities should require scientists to respect ethical standards</li><li>Scientists who had a higher risk perception — that is, who felt that they might experience negative consequences due to research activities — were less likely to believe that scientists should be allowed to do whatever research they want to.</li><li>And much more!</li></ul><p>This paper had several hypotheses and even more results, so I’m going to cut it short there. There are a few things to keep in mind when you think about these results:</p><ul><li>The survey they sent out was answered fully by 444 people. Given that there are way more than 444 nano-scientists, let alone scientists, in the world, these results can’t necessarily be generalized, and probably shouldn’t be.</li><li>82.5% of the people who answered the survey were men. So, when I say that female scientists had a stronger sense of responsibility than men, keep in mind that the results may be a bit biased because of the small number of women who responded.</li><li>Similarly, 2% of the people who responded were Hispanic, and 0.5% were African American. 63.5% identified as White, and 31.8% identified as Asian. In other words, racial diversity was not well represented in this survey, and a more diverse set of survey respondents might lead to different results.</li><li>74% of the people who responded were academic researchers, so there may also be some bias in those comparisons.</li></ul><p>What can we take from this? Well, it probably doesn’t surprise anyone that people’s values affect their opinions on social and ethical issues. However, it should be noted that funding sources seem to impact perceptions of responsibility, in spite of the fact that your funding source isn’t necessarily related to issues of social and ethical responsibility. In addition, “ethical standards” isn’t really defined in this paper, so it probably varies between scientists and between fields. We should make an effort to broadly define “ethics standards” that could apply to all fields, as well as standards for specific areas of science, so that we can mount a united front as policy advocates and incorporate the concept of social and ethics responsibility in science into science funding standards. Although government regulation may not always be able to keep up with scientific advancements, we can all do our part to help policymakers support emerging technology without allowing potentially dangerous advancements to go unchecked.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=da2c1d980c8f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/nanoscientists-and-ethics-paper-of-the-week-june-16th-da2c1d980c8f">Nanoscientists and Ethics — Paper of the Week — June 16th</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Steampunk/MRI Gloves — Paper of the Week — June 6th]]></title>
            <link>https://medium.com/science-communication-policy/steampunk-mri-gloves-paper-of-the-week-june-6th-d8b9b9dad59f?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/d8b9b9dad59f</guid>
            <category><![CDATA[science-education]]></category>
            <category><![CDATA[engineering]]></category>
            <category><![CDATA[health]]></category>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[medicine]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Sat, 29 Dec 2018 16:26:11 GMT</pubDate>
            <atom:updated>2018-12-30T19:44:30.465Z</atom:updated>
            <content:encoded><![CDATA[<h3>Steampunk/MRI Gloves — Paper of the Week — June 6th</h3><p>This week’s paper can be found here: <a href="https://www.nature.com/articles/s41551-018-0233-y">A high-impedance detector-array glove for magnetic resonance imaging of the hand</a></p><p>Note: This paper is behind a paywall on the Nature Biomedical Engineering website. Unfortunately, this is all too common for scientific literature, making it difficult for non-scientists (or even scientists who are not from the same field) to access current research. If you do have access through your institution, I’d encourage you to check it out! If you don’t, I’d encourage you to join the discussion over open access publications that the academic community is currently having. There is also an <a href="https://arxiv.org/pdf/1709.03416.pdf">earlier paper</a> from this lab on a similar topic.</p><p><strong>Welcome back to Paper of the Week!</strong></p><p>We are two for two on papers that I found via Twitter! I’m not sure if that is a good thing, but I can tell you that this is a really cool paper and I am really excited to write about it.</p><p>Note #2: Have a cool paper that you’d like to see me talk about? Thought this paper was cool and want to talk about it? Have questions about STEM, undergrad, grad (haven’t started yet so no promises there), or anything else? Tweet it to me -&gt; <a href="https://twitter.com/jordanbharrod">@jordanbharrod</a></p><p>This week’s paper is titled “A high-impedence detector-array glove for magnetic resonance imaging of the hand” and it comes from New York University’s School of Medicine.</p><p>Why is this paper so cool? Well, for anyone who has ever had an MRI, we know that they are rather large and unbecoming machines that force our bodies to bend to their will (Yes, it’s not that bad, but still). During my undergrad, I volunteered for a bunch of fMRI studies that required me to sit in MRIs for a couple hours every week (it was worth it $$$), so I’ve been in enough different MRI machines to know that, in most clinical cases, there are two rules:</p><ol><li>You can either have high spatial resolution (really detailed images) or high temporal resolution (a lot of images in a short period of time), but not both.</li><li>Don’t move.</li></ol><p>The second rule is the one that really gets to us — the people getting the MRI — because what if you have an itch? Or you have to sneeze? Hopefully, you have the self-control to restrain yourself, because otherwise, you might have to retake that MRI.</p><p>Why is that the case? Well, it has to do with the way that MRI images are taken and reconstructed. If you’re interested in an in-depth review, check out <a href="http://mriquestions.com/index.html">this website</a>, which I found to be really useful when I was taking the medical imaging class that I would later go on to TA. In short, the raw MRI data contains information about where that each data point should be in the reconstructed image, and how bright that data point should be. When you move, the information about where that data point should be is changed, and the reconstructed image does not accurately represent your body.</p><p>This second rule is inconvenient for another reason — MRI is typically used to image tissues (as opposed to bone), and could be used to monitor how tissues move after an injury, which currently can’t be done. This has applications in areas like sports medicine, where muscle injuries are common but can be hard to diagnose.</p><p>But we’re in luck! This detector array glove has the potential to erase that second rule by allowing doctors to image a moving hand during an MRI without the reconstruction problems that we would normally have. The glove uses high-impedence coils (MRIs typically use low impedence) that are attached to your standard glove, which you would then wear into the MRI machine for imaging.</p><p>Why high impedence over low impedence? Let’s go back to how MRI images are taken. When you lay down in an MRI for a scan, the machine uses extremely strong magnets to apply a magnetic field to all of the protons in your body, aligning them. Then, another field is applied, causing protons in different tissues to move away from this alignment to varying degrees. The MRI machine’s detector records the energy each proton releases as they realign with the original magnetic field. The detector is typically made up of those low impedence coils.</p><p>Impendence is a measure of how much resistance to current flow something exhibits, so low impedence coils allow a lot of current to flow. Unfortunately, that current can interfere with the other coils in the MRI, which messes up the resultant image, so the coils can only be arranged in specific designs to prevent that interference. On the other hand, this glove uses high impedence coils, which prevent current flow, and therefore prevent that interference between coils, allowing you to move your hand in the glove without ruining the image.</p><p>Your next question might be: Why does high impedence seem to work better for this application than low impedence? Well, that’s where my expertise runs out. Based on the paper, it seems like the appeal of low impedence coils was energy efficiency — you can use less energy to record the signal for the image. High impedence coils might require more energy, at which point the cost of energy used might not be worth it (or may be unsafe).</p><p>In either case, I thought this paper was very cool, and am looking forward to further developments in the field of MRI that might expand applications towards high resolution images of moving patients for sports medicine!</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d8b9b9dad59f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/steampunk-mri-gloves-paper-of-the-week-june-6th-d8b9b9dad59f">Steampunk/MRI Gloves — Paper of the Week — June 6th</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Dance Me If You Can — Motion Transference using AI]]></title>
            <link>https://medium.com/science-communication-policy/dance-me-if-you-can-motion-transference-using-ai-b58c3cc54c06?source=rss----c10e6d98d139---4</link>
            <guid isPermaLink="false">https://medium.com/p/b58c3cc54c06</guid>
            <category><![CDATA[science-communication]]></category>
            <category><![CDATA[dance]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Jordan Harrod]]></dc:creator>
            <pubDate>Fri, 28 Dec 2018 21:58:25 GMT</pubDate>
            <atom:updated>2018-12-30T19:44:46.397Z</atom:updated>
            <content:encoded><![CDATA[<h3>Dance Me If You Can — Motion Transference using AI</h3><p>This week’s Paper of the Week can be found here: <a href="https://arxiv.org/pdf/1808.07371.pdf">Everybody Dance Now</a></p><p>I don’t know about you, but I’ve never won a dance-off. I wouldn’t consider myself a particularly bad dancer, but the fact of the matter is that there is almost always someone who shows up on the dance floor and starts busting out moves that most of us have only ever seen in choreographed music videos. Now, unlike many of my previous Papers of the Week, this week we are not talking about life or death medical scenarios. As much as we all hate to lose, by the next morning, we return to our normal lives, which presumably do not resemble <em>Footloose</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UVkQoXr7jwsK14A-kqRxrg.png" /><figcaption>Chan et al. 2018</figcaption></figure><p>Instead, we might assuage our dancing deficiencies by watching YouTube choreography or music videos with interesting dance scenes. We might even look into online dance competitions, but only as a spectator, never as a competitor. Until recently, I would wager that most people had come to terms with the current state of their dance skills and were not in a hurry to find a workaround to winning online dance competitions.</p><p>Fortunately for your future online dance competitions, researchers at the University of California, Berkeley, have your back. Using artificial intelligence, they’ve developed a way to transfer the motion of a dancer in a music video directly to you — well, a video of you.</p><p><em>Originally published at </em><a href="http://www.jordanharrod.com/"><em>www.jordanharrod.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b58c3cc54c06" width="1" height="1" alt=""><hr><p><a href="https://medium.com/science-communication-policy/dance-me-if-you-can-motion-transference-using-ai-b58c3cc54c06">Dance Me If You Can — Motion Transference using AI</a> was originally published in <a href="https://medium.com/science-communication-policy">Science + Communication + Policy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>