Bioengineering: A Mechanical Engineering Crossroads
|Assistant Professor Sangbae Kim works on his lab’s current bioinspired project, the robotic cheetah.|
|Photo Credit: M. Scott Brauer|
by Alissa Mallinson
MIT’s Department of Mechanical Engineering has been at the forefront of cutting-edge bioengineering since before its inception, with renowned MIT MechE faculty members like Ioannis V. Yannas developing the first artificial skin, Robert Mann creating the first EMG-controlled prosthetic arm, and Ascher Shapiro and C. Forbes Dewey Jr. uncovering the mysteries of cardiovascular disease.
A short 40 years later, many groundbreaking discoveries have been made that moved the field forward by leaps and bounds – levels of fundamental understanding and technological advancements that could barely have been imagined when MechE first began research activities in bioengineering. Professor Roger Kamm has developed in vitro microfluidic devices into a class of their own, advancing the understanding of cancer metastasis and paving the way for life-changing drug delivery. Associate Professor Anette “Peko” Hosoi and Assistant Professor Amos Winter have created robots that burrow and anchor in the ocean inspired by the natural burrowing of clams, and Professor Harry Asada has developed light-activated artificial muscles for robotic movement. Professor Peter So has developed super microscopes that allow 3D single-cell visualizations of ex vivo animal organs, and Associate Professor Domitilla Del Vecchio is designing feedback controllers in cells to realize biological operational amplifiers.
Bioengineering is a field with tremendous opportunities for impacting human health. It is rich with mystery and possibility, and has opened the flood gates for exploration and exploitation, for engineers in particular, who are now applying biological principles to mechanical design and, conversely, mechanical engineering principles to the understanding of biology, advancing both fields dramatically in the process. “Engineers reveal nature’s principles by building,” says Professor Ian W. Hunter. “The act of building forces you to pose questions you might not have asked yourself otherwise, and bioengineering is no exception.”
Our faculty are focused on multiple facets of bioengineering, including bio-inspired design, bioinstrumentation, biomedical devices, and biomolecular systems and control. Major new trends continue to arise from and influence their research, such as bio-integrated engineering and consumer empowerment.
Although recently coined as “bio-inspired design,” the idea of applying nature’s refined designs to man-made creations is one that can be traced back to the time of Leonardo da Vinci, whose observations of biology informed much of his work. Several noteworthy inventions were created in much this same way: the airplane, from studying the flight of winged animals, and Velcro, from noticing burrs attached to dog hair, to name two well-known examples.
These days, that same principle is being used to design more efficient, flexible robots, such as Professor Sangbae Kim’s robotic cheetah or Professor Hosoi’s robotic snail.
“Technology maturity has reached a certain threshold at this point that encourages people to apply the principles we learn from animals and nature,” says Kim.
“We’re not copying biological systems,” adds Hosoi, “but rather understanding their physical principles and applying those to engineering design.”
As the field of bio-inspired design progresses, a new field of synthetic biology is unfolding alongside it. “Synthetic biology – such as producing methanol from something like switch grass – would be a great way to produce drugs,” says Dewey, “because it’s very efficient and reproducible.”
That, in turn, leads to even further development. Professor Kamm explains:
“Through recent advances in regenerative medicine and synthetic biology, the possibility of creating multicellular biological machines has come onto the horizon. Our vision is that these fully biological, cell-based systems will someday function side-by-side with traditional machines created from inert materials, opening up an exciting new opportunity in which biological and inert materials function together seamlessly to perform their designed function.”
According to Kamm, one example of this new field, termed bio-integrated engineering, would be the ability to produce “hyper-organs.” Consider an ultra-sensitive nose that “smells,” then feeds the information it gathers via a network of neurons to a conventional CPU that analyzes it (see Faculty Research: Professor Harry Asada).
Professor Hunter and Dr. Brian Hemond’s micro mass spectrometer is representative of another upcoming bioengineering trend: consumer empowerment. “People are getting tired of relying on experts for everything,” says Hunter. “They want to empower themselves with the ability to measure things on their own. In the past, people would have their blood pressure measured at a doctor’s office, but now there’s no reason you can’t do this at home. But what else would people like to measure? You could have portable diagnostic devices in the home that would analyze blood samples, or you could even look through the tissue to analyze blood, ultimately bringing the instrument to the specimen instead of the other way around. I see the miniaturization of advanced diagnostic instruments as an important trend of the future.”
An inextricable part of that trend is lowering the cost of manufacturing, as well as incorporating ways of analyzing the data, he says.
On the flip side of consumer empowerment is expert empowerment, another trend that bioinstrumentation has a big hand in developing. Scientists and medical doctors also need smaller, more precise instruments to get the job done. They need to get closer, less invasively, and more easily if they are to continue necessary progress. To this end, Professor So has exploited advances in photonics to develop game-changing high-throughput/high-content 3D imaging bioinstrumentation.
“Basic biology is important,” says So, “and advanced technology enables the study of something much more basic than you could study before. If you have a better understanding of basic biology, you can make positive contributions to human health, food production, environmental protection, and other important sociological problems that are biological in origin.”
In the biomedical device field, lowering costs is a concern as well. As high health care costs continue to rise, people will continue to look for cheaper, faster, and more accurate health care.
“There’s going to be this large push for people to look for ways to cut health care costs, and you’ll see medical devices actually boom because of that,” predicts Professor Douglas Hart. “If we can make diagnostics more accurate, easier, and cheaper, then why wouldn’t we?”
But it’s not just diagnostics that benefit from improved devices. Quality of life could be an area of significant improvement as well. Advanced surgical devices (such as the flexural laparoscopic grasper designed by MechE students), for example, can cut down on surgery and recovery time as well as decrease likelihood of infection and speed up wound healing time (see Alumni Spotlight on Danielle Zurovcik PhD ‘12), or monitor patient vitals in the comfort of their own home (exemplified by the Sombus Sleep Shirt developed by Rest Devices, a medical device company started by three MechE students in Professor Alex Slocum’s 2.75 course).
On the research side, 3D in vitro microfluidic devices, such as those first developed in the laboratory of Professor Kamm, are helping our faculty discover effective ways of preventing the spread of disease – for example, cancer metastasis in Professor Kamm’s case – and enabling engineers to unveil new methods of drug delivery – for example, through leaky endothelial vessels (see Faculty Research: Professor Roger Kamm).
Biomolecular Control Systems
Biomolecular control systems may not be officially labeled as “biomedical,” but their potential for controlling cells is quickly being recognized as a crucial element of solving several societal problems, including energy and health care. Through biological control systems, engineers are able to use and develop control theory to craft modular designs of biomolecular circuits that can be inserted into living cells to control their behavior.
“The ability to control cell behavior,” explains Associate Professor Del Vecchio, “has set the stage for groundbreaking applications ranging from biofuels and biosensing to molecular computing and targeted drug delivery. We envision a near future in which programmable cells will transform waste into energy and kill cancer cells in the bloodstream.”
Biomolecular circuitry is limited by its size and complexity, and for the most part current research is focused on establishing design principles to overcome these limitations and enable larger, more complex circuits.
Some mechanical engineers such as Professor Jean-Jacques Slotine are utilizing mathematical principles and abstract control theory to identify the characteristics of a network that would make it more or less easy to control, and determine which nodes and locations in the network need to be controlled to make it all work.
The Sum of its Parts
“We were scratching the surface of bioengineering for a long time with something slightly beyond taxonomy,” says Dewey. “Obviously some brilliant things have already come along, but as we learn more and more, new opportunities continue to arise. The detective work to get through that is daunting, but once you get there, the question is, ‘how do you use it?’”
The answer must include extensive cross-disciplinary collaboration to fulfill its potential. The best solutions and discoveries utilize and integrate the many facets of mechanical engineering: mechanics, manufacturing, design and prototyping, nanotechnology, computation, robotics, and controls. As is often the case with mechanical engineering, it’s the sum of its parts, working together, that will change lives for the better.
Professor Anette Hosoi
Taking their lead from nature, MechE faculty members like Professor Peko Hosoi and her team are developing significantly improved robotic devices to tackle modern challenges. With a background in physics and fluid mechanics, Hosoi gets her inspiration from lower-level organisms. “There are a couple of key things we look for,” she says. “Simplicity, a system grounded in mechanics, and significantly better performance than can already be achieved in engineered systems.”
With that in mind, she led her research team in an effort to create the most versatile crawler and landed upon the snail – “nature’s ultimate all-terrain vehicle,” as she calls it. Upon further investigation, they found that when a snail crawls, a trail of slime is left behind that allows it to climb up almost anything. This yield-stress fluid is in a solid-like state during the climb, sticking snails to the wall like glue, but when they start to shear it with the bottom of their foot, it begins to flow like a liquid. “But if you only have one foot and you are sitting on a layer of glue, how do you move?” asked Hosoi. “The answer isn’t obvious.”
The Hosoi team discovered that snails’ muscles contract to create a wave of compression along the bottom of the foot, and invented a way to utilize this naturally occurring technique to design a highly advanced crawler. In their design, a non-Newtonian fluid reacts with two solid pads on the main body – one consisting of a series of small pads that act as one large pad, and a single smaller pad. The difference in force based on varying sizes breaks the symmetry and allows one pad to stick and the other pad to slide, thus moving the RoboSnail forward.
The Hosoi team’s invention is currently being used in the oil industry as a way to move instruments along pipes in “down-hole environments”; however, RoboSnail can be useful on any terrain.
|Professor Hosoi’s RoboSnail.|
|Photo courtesy of Professor Hosoi|
Professor Sangbae Kim
Assistant Professor Sangbae Kim is also inspired by nature’s ideal solutions. He and his team have been building a robotic cheetah in their lab and recently completed the design and prototype of a robotic earthworm. Like Hosoi, they aren’t interested in unraveling the mystery of intelligence but rather in understanding the mobility of accomplished animals and borrowing their successful characteristics to create advanced mobility in robots.
“Our core motivation is to understand how biological systems are designed,” says Kim. “People make a lot of assumptions about what animals are designed and optimized for. We’re fascinated by that – people think it’s the ideal design, but we’re not sure. Our research is centered around the idea of testing and proving those hypotheses, and using those findings to form the basis for a new perspective.”
The team’s newest perspective came from analyzing a simple earthworm. Similar to the Robosnail’s wave of contractions, the earthworm also moves forward by squeezing and stretching its muscles, a mechanism called peristalsis. Kim’s resulting Meshworm is made of an artificial muscle constructed from nickel and titanium wire that stretches and contracts multiple segments of its body using heat from a small current. Kim and his team developed algorithms to carefully control the wire’s heating and cooling, directing the worm’s movement.
Professor Ian Hunter
When Professor Ian Hunter initially had the idea for a needleless jet injector that could provide various dose amounts of differing drugs at multiple depths, he doubted if it would even be possible. But years later, the injector, uniquely controlled by a Lorentz-force actuator, is an elegant and easy option for otherwise difficult drug delivery. Needleless injection isn’t a new idea, but Hunter’s creativity, combined with the vision and expertise of his team – Dr. Cathy Hogan, Dr. Andrew Taberner, and recent PhD graduates Dr. Brian Hemond and Dr. Adam Wahab – has hoisted it to a new level. The team is the first to use a lab-made Lorentz-force actuator to control velocity, volume, and pressure to deliver drugs at a rate equivalent to the speed of sound in air. It can deliver drugs superficially into skin, deep into muscle, through the eye into the retina, through the tempanic membrane into the middle ear, and even into interstitial fluid. The device – which, at 10 to 20 milliseconds per dose (anywhere from 105 micro liters to 500 micro liters), is speedy and almost silent – can also utilize its wide bandwidth to vibrate drugs in solid powder form to create a fluidized drug, thus solving the cold chain problem of delivering refrigerated liquid vaccines to third world countries. Receiving the 10 to 100 Joules needed to power a single-dose delivery from a small lithium polymer battery, the device is also bidirectional, allowing for the option of sucking biomaterials out of the body – including DNA and proteins. The injector uses a 32-bit microcontroller for controlling its operation and a digital nonlinear feedback control system running at up to 100,000 samples per second. The newest version of Hunter’s injector includes a double Halbach magnet array designed by recent PhD graduate Dr. Brian Ruddy. It is expected to deliver a dose of drugs with a precision of less than one microliter.
Professor Peter So
The question of what differentiates alternate paths of diseases, or why one type of biomaterial more successfully regenerates tissue than another is directed by many cellular and molecular factors that are very difficult to decipher.
Professor Peter So’s team is getting to the answers by mapping tissue morphological structures and biochemical organizations of small animal organs with subcellular resolution. The high-throughput/high-content 3D imaging bioinstrumentation they developed as a result of this focus is based on high-speed multi-photon microscopy that enables them to study tissue structures that span five orders of magnitude in scale.
Instead of exciting fluorescence using a single blue photon, So’s team uses a lower-energy infrared light to ensure that the probability of photointeraction is limited to a volume on the order of one femtoliter. Outside of this spot, the photon flux is lower and there is no excitation, thus providing unique 3D resolution.
Professor Douglas Hart
Professor Douglas Hart’s development of the first 3D ear canal scanner for making perfect-fit hearing aids was actually the result of a happy accident in an auto shop. While measuring oil-film thickness and temperature in seals on the cylinder wall of engines, one of Hart’s students ran into a problem: He couldn’t get the oil thickness out of the equations. Instead of getting frustrated, Hart took advantage of the problem. “We started realizing that if we could measure the thickness of the liquid that we could also get a 3D image of it,” says Hart.
Currently the only 3D ear canal scanner available, the portable Lantos Scanner provides safe, comfortable and fast mapping of the ear canal for perfect hearing aid fits, as well as custom ear plugs and internal headphones.
The soft membrane at the end of the scanner has a visible odoscope tip that is guided deep into the ear canal by a video feed. Once inserted the membrane is filled with a water-based optical dye, expanding the membrane until it conforms perfectly to the patient’s unique inner ear. Using a dual wave-length algorithm, the scanner takes thousands of 2D digital images as it exits the ear and stitches them together in real time to create a highly accurate 3D image of the ear. The scanner can even measure canal wall elasticity by varying the pressure inside the membrane and recording the results. An on-site laptop computer processes all the data and sends it directly to the manufacturer.
“Ears are very small and don’t have texture to them, so you can’t use stereoscopic imaging to get images of them, nor structured light systems because of skin translucency,” explains Hart. “People have tried interferometry and other similar ideas, but that’s too sensitive and expensive. What we have designed is one of a kind, utilizing color absorption ratios to determine distance.”
Because of the scanner’s incredible accuracy, not only can it notably improve patient comfort and hearing aid quality, but it also significantly cuts down on manufacturing cost by eliminating the need to remake hearing aids that weren’t fitted correctly the first time.
Professor Alexander Slocum
Professor Alexander Slocum has been guiding senior undergraduates as well as graduate students interested in biomedical device design since 2004, when he founded Course 2.75 upon the famous MIT credo mens et manus (“mind and hand”). The 14-week course is focused on biomedical device design projects that match teams of students with Boston-area clinicians. Each team of three to five students works on a real problem, from brainstorming and designing a proof-of-concept device, to building and testing a prototype – all along the way constantly challenged by Slocum to identify risks and viable countermeasures to the design and its production.
One look at the list of devices and start-up companies that have spun off from the work done by 2.75 students, and it’s easy to see that Slocum’s hands-on approach to the course works. For example, Robopsy, a robotic device to assist radiologists during percutaneous tumor biopsies, won the 2007 $100K Entrepreneurship Competition and the first-place award at the 2008 ASME Innovation Showcase. In 2011, the Somnus™ Sleep Shirt, which monitors patient respiration at home, won a prize at the prestigious Three-in-Five Competition at the Design of Medical Devices Conference. Upon graduation, the shirt’s creators formed Rest Devices (http://www.restdevices.com) to commercialize their technology and secured $500,000 in angel financing. Following both a clinical and a consumer path, they are simultaneously developing the Somnus sleep monitoring shirt as an alternative to in-hospital sleep studies while preparing to launch an infant monitoring onesie. Several other devices from the class have been licensed and are expected to be in production soon.
Slocum’s former TA for the class, Dr. Nevan Hanumara, is now a post-doc in his lab, working to expand the course and create an industry outreach program that gets more of the course’s products into production.
Professor Domitilla Del Vecchio
The term “network” has become a standard metaphor for describing the system of arteries and synapases and other areas of flow and synergy in the body, but there’s another, less common image that illustrates the connectivity perhaps even better: circuitry. As with electrical circuits, you get effects similar to impedance called “retroactivity,” says Professor Domitilla Del Vecchio, who was one of the first to apply control theory ideas to biomolecular design systems that are impervious to such impedance.
“The main problem is how one can use the specific mechanism you have in biological systems to modify or create new control techniques that are useful in this new domain,” says Del Vecchio.
Del Vecchio’s team is currently focused on the development of biomolecular feedback circuits that are robust to retroactivity and function like operational amplifiers in electronics. “Biological components are already there,” says Del Vecchio, “so our focus is on making the ensemble of these parts suitable for modular design, which will enable the creation of complex new functionalities.”
Professor Jean-Jacques Slotine
One area of Professor Jean-Jacques Slotine’s focus is the theory of biological control systems. He’s researching ways to control and exploit synchronization mechanisms in neurons and cells, and utilizing mathematical principles to answer the question of which nodes and locales in a biological circuit network need to be controlled to gain control of the entire network. Now that biological models are precise enough to be controlled, says Slotine, we can begin to develop circuit diagrams and dynamical systems that might be able to do the job of controlling the large, nonlinear, and complex networks that exist in biology.
His recent work with colleagues points toward the origin and characteristics of a network that would allow for easy control. He has discovered that there are specific characteristics that make a biological network easy or difficult to control, and that more connections equate to greater control of the network as a whole.
Although his work is very theoretical at this stage, Slotine suspects that such fundamental knowledge about how to control biological networks could be very useful in the future for drug delivery and synthetic biology.