Tag Archives: knowledge
#434648 The Pediatric AI That Outperformed ...
Training a doctor takes years of grueling work in universities and hospitals. Building a doctor may be as easy as teaching an AI how to read.
Artificial intelligence has taken another step towards becoming an integral part of 21st-century medicine. New research out of Guangzhou, China, published February 11th in Nature Medicine Letters, has demonstrated a natural-language processing AI that is capable of out-performing rookie pediatricians in diagnosing common childhood ailments.
The massive study examined the electronic health records (EHR) from nearly 600,000 patients over an 18-month period at the Guangzhou Women and Children’s Medical Center and then compared AI-generated diagnoses against new assessments from physicians with a range of experience.
The verdict? On average, the AI was noticeably more accurate than junior physicians and nearly as reliable as the more senior ones. These results are the latest demonstration that artificial intelligence is on the cusp of becoming a healthcare staple on a global scale.
Less Like a Computer, More Like a Person
To outshine human doctors, the AI first had to become more human. Like IBM’s Watson, the pediatric AI leverages natural language processing, in essence “reading” written notes from EHRs not unlike how a human doctor would review those same records. But the similarities to human doctors don’t end there. The AI is a machine learning classifier (MLC), capable of placing the information learned from the EHRs into categories to improve performance.
Like traditionally-trained pediatricians, the AI broke cases down into major organ groups and infection areas (upper/lower respiratory, gastrointestinal, etc.) before breaking them down even further into subcategories. It could then develop associations between various symptoms and organ groups and use those associations to improve its diagnoses. This hierarchical approach mimics the deductive reasoning human doctors employ.
Another key strength of the AI developed for this study was the enormous size of the dataset collected to teach it: 1,362,559 outpatient visits from 567,498 patients yielded some 101.6 million data points for the MLC to devour on its quest for pediatric dominance. This allowed the AI the depth of learning needed to distinguish and accurately select from the 55 different diagnosis codes across the various organ groups and subcategories.
When comparing against the human doctors, the study used 11,926 records from an unrelated group of children, giving both the MLC and the 20 humans it was compared against an even playing field. The results were clear: while cohorts of senior pediatricians performed better than the AI, junior pediatricians (those with 3-15 years of experience) were outclassed.
Helping, Not Replacing
While the research used a competitive analysis to measure the success of the AI, the results should be seen as anything but hostile to human doctors. The near future of artificial intelligence in medicine will see these machine learning programs augment, not replace, human physicians. The authors of the study specifically call out augmentation as the key short-term application of their work. Triaging incoming patients via intake forms, performing massive metastudies using EHRs, providing rapid ‘second opinions’—the applications for an AI doctor that is better-but-not-the-best are as varied as the healthcare industry itself.
That’s only considering how artificial intelligence could make a positive impact immediately upon implementation. It’s easy to see how long-term use of a diagnostic assistant could reshape the way modern medical institutions approach their work.
Look at how the MLC results fit snugly between the junior and senior physician groups. Essentially, it took nearly 15 years before a physician could consistently out-diagnose the machine. That’s a decade and a half wherein an AI diagnostic assistant would be an invaluable partner—both as a training tool and a safety measure. Likewise, on the other side of the experience curve you have physicians whose performance could be continuously leveraged to improve the AI’s effectiveness. This is a clear opportunity for a symbiotic relationship, with humans and machines each assisting the other as they mature.
Closer to Us, But Still Dependent on Us
No matter the ultimate application, the AI doctors of the future are drawing nearer to us step by step. This latest research is a demonstration that artificial intelligence can mimic the results of human deductive reasoning even in some of the most complex and important decision-making processes. True, the MLC required input from humans to function; both the initial data points and the cases used to evaluate the AI depended on EHRs written by physicians. While every effort was made to design a test schema that removed any indication of the eventual diagnosis, some “data leakage” is bound to occur.
In other words, when AIs use human-created data, they inherit human insight to some degree. Yet the progress made in machine imaging, chatbots, sensors, and other fields all suggest that this dependence on human input is more about where we are right now than where we could be in the near future.
Data, and More Data
That near future may also have some clear winners and losers. For now, those winners seem to be the institutions that can capture and apply the largest sets of data. With a rapidly digitized society gathering incredible amounts of data, China has a clear advantage. Combined with their relatively relaxed approach to privacy, they are likely to continue as one of the driving forces behind machine learning and its applications. So too will Google/Alphabet with their massive medical studies. Data is the uranium in this AI arms race, and everyone seems to be scrambling to collect more.
In a global community that seems increasingly aware of the potential problems arising from this need for and reliance on data, it’s nice to know there’ll be an upside as well. The technology behind AI medical assistants is looking more and more mature—even if we are still struggling to find exactly where, when, and how that technology should first become universal.
Yet wherever we see the next push to make AI a standard tool in a real-world medical setting, I have little doubt it will greatly improve the lives of human patients. Today Doctor AI is performing as well as a human colleague with more than 10 years of experience. By next year or so, it may take twice as long for humans to be competitive. And in a decade, the combined medical knowledge of all human history may be a tool as common as a stethoscope in your doctor’s hands.
Image Credit: Nadia Snopek / Shutterstock.com Continue reading
#434643 Sensors and Machine Learning Are Giving ...
According to some scientists, humans really do have a sixth sense. There’s nothing supernatural about it: the sense of proprioception tells you about the relative positions of your limbs and the rest of your body. Close your eyes, block out all sound, and you can still use this internal “map” of your external body to locate your muscles and body parts – you have an innate sense of the distances between them, and the perception of how they’re moving, above and beyond your sense of touch.
This sense is invaluable for allowing us to coordinate our movements. In humans, the brain integrates senses including touch, heat, and the tension in muscle spindles to allow us to build up this map.
Replicating this complex sense has posed a great challenge for roboticists. We can imagine simulating the sense of sight with cameras, sound with microphones, or touch with pressure-pads. Robots with chemical sensors could be far more accurate than us in smell and taste, but building in proprioception, the robot’s sense of itself and its body, is far more difficult, and is a large part of why humanoid robots are so tricky to get right.
Simultaneous localization and mapping (SLAM) software allows robots to use their own senses to build up a picture of their surroundings and environment, but they’d need a keen sense of the position of their own bodies to interact with it. If something unexpected happens, or in dark environments where primary senses are not available, robots can struggle to keep track of their own position and orientation. For human-robot interaction, wearable robotics, and delicate applications like surgery, tiny differences can be extremely important.
Piecemeal Solutions
In the case of hard robotics, this is generally solved by using a series of strain and pressure sensors in each joint, which allow the robot to determine how its limbs are positioned. That works fine for rigid robots with a limited number of joints, but for softer, more flexible robots, this information is limited. Roboticists are faced with a dilemma: a vast, complex array of sensors for every degree of freedom in the robot’s movement, or limited skill in proprioception?
New techniques, often involving new arrays of sensory material and machine-learning algorithms to fill in the gaps, are starting to tackle this problem. Take the work of Thomas George Thuruthel and colleagues in Pisa and San Diego, who draw inspiration from the proprioception of humans. In a new paper in Science Robotics, they describe the use of soft sensors distributed through a robotic finger at random. This placement is much like the constant adaptation of sensors in humans and animals, rather than relying on feedback from a limited number of positions.
The sensors allow the soft robot to react to touch and pressure in many different locations, forming a map of itself as it contorts into complicated positions. The machine-learning algorithm serves to interpret the signals from the randomly-distributed sensors: as the finger moves around, it’s observed by a motion capture system. After training the robot’s neural network, it can associate the feedback from the sensors with the position of the finger detected in the motion-capture system, which can then be discarded. The robot observes its own motions to understand the shapes that its soft body can take, and translate them into the language of these soft sensors.
“The advantages of our approach are the ability to predict complex motions and forces that the soft robot experiences (which is difficult with traditional methods) and the fact that it can be applied to multiple types of actuators and sensors,” said Michael Tolley of the University of California San Diego. “Our method also includes redundant sensors, which improves the overall robustness of our predictions.”
The use of machine learning lets the roboticists come up with a reliable model for this complex, non-linear system of motions for the actuators, something difficult to do by directly calculating the expected motion of the soft-bot. It also resembles the human system of proprioception, built on redundant sensors that change and shift in position as we age.
In Search of a Perfect Arm
Another approach to training robots in using their bodies comes from Robert Kwiatkowski and Hod Lipson of Columbia University in New York. In their paper “Task-agnostic self-modeling machines,” also recently published in Science Robotics, they describe a new type of robotic arm.
Robotic arms and hands are getting increasingly dexterous, but training them to grasp a large array of objects and perform many different tasks can be an arduous process. It’s also an extremely valuable skill to get right: Amazon is highly interested in the perfect robot arm. Google hooked together an array of over a dozen robot arms so that they could share information about grasping new objects, in part to cut down on training time.
Individually training a robot arm to perform every individual task takes time and reduces the adaptability of your robot: either you need an ML algorithm with a huge dataset of experiences, or, even worse, you need to hard-code thousands of different motions. Kwiatkowski and Lipson attempt to overcome this by developing a robotic system that has a “strong sense of self”: a model of its own size, shape, and motions.
They do this using deep machine learning. The robot begins with no prior knowledge of its own shape or the underlying physics of its motion. It then repeats a series of a thousand random trajectories, recording the motion of its arm. Kwiatkowski and Lipson compare this to a baby in the first year of life observing the motions of its own hands and limbs, fascinated by picking up and manipulating objects.
Again, once the robot has trained itself to interpret these signals and build up a robust model of its own body, it’s ready for the next stage. Using that deep-learning algorithm, the researchers then ask the robot to design strategies to accomplish simple pick-up and place and handwriting tasks. Rather than laboriously and narrowly training itself for each individual task, limiting its abilities to a very narrow set of circumstances, the robot can now strategize how to use its arm for a much wider range of situations, with no additional task-specific training.
Damage Control
In a further experiment, the researchers replaced part of the arm with a “deformed” component, intended to simulate what might happen if the robot was damaged. The robot can then detect that something’s up and “reconfigure” itself, reconstructing its self-model by going through the training exercises once again; it was then able to perform the same tasks with only a small reduction in accuracy.
Machine learning techniques are opening up the field of robotics in ways we’ve never seen before. Combining them with our understanding of how humans and other animals are able to sense and interact with the world around us is bringing robotics closer and closer to becoming truly flexible and adaptable, and, eventually, omnipresent.
But before they can get out and shape the world, as these studies show, they will need to understand themselves.
Image Credit: jumbojan / Shutterstock.com Continue reading
#433884 Designer Babies, and Their Babies: How ...
As if stand-alone technologies weren’t advancing fast enough, we’re in age where we must study the intersection points of these technologies. How is what’s happening in robotics influenced by what’s happening in 3D printing? What could be made possible by applying the latest advances in quantum computing to nanotechnology?
Along these lines, one crucial tech intersection is that of artificial intelligence and genomics. Each field is seeing constant progress, but Jamie Metzl believes it’s their convergence that will really push us into uncharted territory, beyond even what we’ve imagined in science fiction. “There’s going to be this push and pull, this competition between the reality of our biology with its built-in limitations and the scope of our aspirations,” he said.
Metzl is a senior fellow at the Atlantic Council and author of the upcoming book Hacking Darwin: Genetic Engineering and the Future of Humanity. At Singularity University’s Exponential Medicine conference last week, he shared his insights on genomics and AI, and where their convergence could take us.
Life As We Know It
Metzl explained how genomics as a field evolved slowly—and then quickly. In 1953, James Watson and Francis Crick identified the double helix structure of DNA, and realized that the order of the base pairs held a treasure trove of genetic information. There was such a thing as a book of life, and we’d found it.
In 2003, when the Human Genome Project was completed (after 13 years and $2.7 billion), we learned the order of the genome’s 3 billion base pairs, and the location of specific genes on our chromosomes. Not only did a book of life exist, we figured out how to read it.
Jamie Metzl at Exponential Medicine
Fifteen years after that, it’s 2018 and precision gene editing in plants, animals, and humans is changing everything, and quickly pushing us into an entirely new frontier. Forget reading the book of life—we’re now learning how to write it.
“Readable, writable, and hackable, what’s clear is that human beings are recognizing that we are another form of information technology, and just like our IT has entered this exponential curve of discovery, we will have that with ourselves,” Metzl said. “And it’s intersecting with the AI revolution.”
Learning About Life Meets Machine Learning
In 2016, DeepMind’s AlphaGo program outsmarted the world’s top Go player. In 2017 AlphaGo Zero was created: unlike AlphaGo, AlphaGo Zero wasn’t trained using previous human games of Go, but was simply given the rules of Go—and in four days it defeated the AlphaGo program.
Our own biology is, of course, vastly more complex than the game of Go, and that, Metzl said, is our starting point. “The system of our own biology that we are trying to understand is massively, but very importantly not infinitely, complex,” he added.
Getting a standardized set of rules for our biology—and, eventually, maybe even outsmarting our biology—will require genomic data. Lots of it.
Multiple countries already starting to produce this data. The UK’s National Health Service recently announced a plan to sequence the genomes of five million Britons over the next five years. In the US the All of Us Research Program will sequence a million Americans. China is the most aggressive in sequencing its population, with a goal of sequencing half of all newborns by 2020.
“We’re going to get these massive pools of sequenced genomic data,” Metzl said. “The real gold will come from comparing people’s sequenced genomes to their electronic health records, and ultimately their life records.” Getting people comfortable with allowing open access to their data will be another matter; Metzl mentioned that Luna DNA and others have strategies to help people get comfortable with giving consent to their private information. But this is where China’s lack of privacy protection could end up being a significant advantage.
To compare genotypes and phenotypes at scale—first millions, then hundreds of millions, then eventually billions, Metzl said—we’re going to need AI and big data analytic tools, and algorithms far beyond what we have now. These tools will let us move from precision medicine to predictive medicine, knowing precisely when and where different diseases are going to occur and shutting them down before they start.
But, Metzl said, “As we unlock the genetics of ourselves, it’s not going to be about just healthcare. It’s ultimately going to be about who and what we are as humans. It’s going to be about identity.”
Designer Babies, and Their Babies
In Metzl’s mind, the most serious application of our genomic knowledge will be in embryo selection.
Currently, in-vitro fertilization (IVF) procedures can extract around 15 eggs, fertilize them, then do pre-implantation genetic testing; right now what’s knowable is single-gene mutation diseases and simple traits like hair color and eye color. As we get to the millions and then billions of people with sequences, we’ll have information about how these genetics work, and we’re going to be able to make much more informed choices,” Metzl said.
Imagine going to a fertility clinic in 2023. You give a skin graft or a blood sample, and using in-vitro gametogenesis (IVG)—infertility be damned—your skin or blood cells are induced to become eggs or sperm, which are then combined to create embryos. The dozens or hundreds of embryos created from artificial gametes each have a few cells extracted from them, and these cells are sequenced. The sequences will tell you the likelihood of specific traits and disease states were that embryo to be implanted and taken to full term. “With really anything that has a genetic foundation, we’ll be able to predict with increasing levels of accuracy how that potential child will be realized as a human being,” Metzl said.
This, he added, could lead to some wild and frightening possibilities: if you have 1,000 eggs and you pick one based on its optimal genetic sequence, you could then mate your embryo with somebody else who has done the same thing in a different genetic line. “Your five-day-old embryo and their five-day-old embryo could have a child using the same IVG process,” Metzl said. “Then that child could have a child with another five-day-old embryo from another genetic line, and you could go on and on down the line.”
Sounds insane, right? But wait, there’s more: as Jason Pontin reported earlier this year in Wired, “Gene-editing technologies such as Crispr-Cas9 would make it relatively easy to repair, add, or remove genes during the IVG process, eliminating diseases or conferring advantages that would ripple through a child’s genome. This all may sound like science fiction, but to those following the research, the combination of IVG and gene editing appears highly likely, if not inevitable.”
From Crazy to Commonplace?
It’s a slippery slope from gene editing and embryo-mating to a dystopian race to build the most perfect humans possible. If somebody’s investing so much time and energy in selecting their embryo, Metzl asked, how will they think about the mating choices of their children? IVG could quickly leave the realm of healthcare and enter that of evolution.
“We all need to be part of an inclusive, integrated, global dialogue on the future of our species,” Metzl said. “Healthcare professionals are essential nodes in this.” Not least among this dialogue should be the question of access to tech like IVG; are there steps we can take to keep it from becoming a tool for a wealthy minority, and thereby perpetuating inequality and further polarizing societies?
As Pontin points out, at its inception 40 years ago IVF also sparked fear, confusion, and resistance—and now it’s as normal and common as could be, with millions of healthy babies conceived using the technology.
The disruption that genomics, AI, and IVG will bring to reproduction could follow a similar story cycle—if we’re smart about it. As Metzl put it, “This must be regulated, because it is life.”
Image Credit: hywards / Shutterstock.com Continue reading