Prosthetic Sculptures Duplicate Faces of Wounded U.S. Soldiers
[Sorry, the video for this story has expired, but you can still read the transcript below. ]
SUSAN DENTZER: Duplicating or replacing a human face is a really complicated problem. Exactly what is the problem?
DAVID HANSON: The problem with the repairing of the human face is the complexity of the organ structure. You have tissues, facial tissues, the muscles, the mucosa, an incredible intersection of complex functions in the face. And not only does it have to maintain the integrity of the sinus and digestive tract, the breathing pathways, you also have to achieve the cosmetic function. We are social creatures and the face has to look good and it has to move well too or else it can be disconcerting.
You have many types of tissues and organs that come together and do a lot of very important human activities. Breathing is done through the face, eating, as well as the social functions of the face. The face has to look good. It has to move well also. If it doesn’t move it’s disconcerting.
SUSAN DENTZER: And nature does a really great job of that. How good are we at doing it, as human beings trying to recreate what nature already achieved?
DAVID HANSON: In robotics, we’re pretty good at achieving that, but combining robotics with an injured face to restore facial expressions is extremely challenging because you have to maintain those biological functions. You have to allow the face to breathe and eat and you have to prevent drooling and this kind of thing. So combining the technologies with the face is very challenging.
SUSAN DENTZER: How would one think about then taking somebody’s whose face had been mostly blown away and putting a robotic prosthesis, a face prosthesis, on top of it?
DAVID HANSON: To restore the cosmetic appearance of an injured soldier’s face, you would need to fit the robotics to the face, contain all of the power systems, actuators, that is motors and sensors, within the appliance and have the appliance be easily maintained, biocompatible, compatible with your tissues and mimic the soldier’s existing facial expressions so the soldier would have to be able to control that by, by thinking. It’s a lot to put into one small package.
SUSAN DENTZER: But you think doable.
DAVID HANSON: Eminently doable. The key is to reduce the power consumption of the expressive materials which is something we focused on with the robots that we’re building, because we want these to walk around and interact with us. To do that you have to be able to power it off of batteries. So tuning the materials, it’s a golden age in material science, so being able to tune the material properties for low power expressions is very doable.
SUSAN DENTZER: Let's talk about how you built these robots. And let's start with the skin. Tell me about this incredible substance that you made, what it's called, how you made it.
DAVID HANSON: We call it Frubber, for face rubber. It's mildly whimsical. But the essence of the material is that we structure pores, or chambers, through the material so that it can stretch very easily with very low power but also compress and that compression makes it move more like facial soft tissue. When you touch the material it feels like human tissue because we can structure those chambers, we're able to get skin-like epidermis and then structure human-like deep tissues as well. That results in a more life-like expression that consumes less than 1/20th the power of a solid rubber material. So if you use a medical grade prosthetic, silicone for example, it would require 20 times the energy to move into expressions.
SUSAN DENTZER: You actually have electrical current flowing through it. How does that work?
DAVID HANSON: Well, we have small motors that connect to linkages in the skin and those motors pull the skin into facial expression. We've been experimenting with embedding the motors in the material. For prosthesis it would be important that the motors are entirely embedded in the prosthesis, but then you have a nice model for that which is the human face. There the motors are muscles, so here we're beginning to emulate the performance of muscles. The best materials, the best motors in the world, to behave like muscles, are electroactive polymer actuators or EAP, artificial muscles. They are not quite ready to embed in prosthetics or in facial robots, but there have been some very promising experiments that we've been doing with the Jet Propulsion Laboratory and with the Stanford Research Institute as well. SRI has very promising artificial muscles. We believe that lowering the power requirements of the skin material and boosting the performance of artificial muscles means you'll have a convergence in these technologies which will enable the, the facial prosthetics.
Bringing technology together
SUSAN DENTZER: You went to this workshop sponsored under the umbrella of the Defense Advanced Research Projects Agency, or DARPA, and you heard about all of these wounded soldiers who have lost almost all or major parts of their faces. Currently there are more than 100, and certainly will be many more before this conflict in Iraq is over and future conflicts after that. What did you think to yourself as you heard the discussions over how these people could be helped?
DAVID HANSON: I thought that we are in a time of great convergence and that we could solve a great deal of human suffering if we could push the revolutions of biotechnology and tissue engineering and robotics together - to better administer to wounded soldiers and civilians as well. There are many more civilians with these kinds of injuries actually in the world and they'll benefit from this convergence of technology.
I felt that I was at a turning point in history, the history of medicine where we could begin to administer medicine with robotics, and that that could represent a full future for addressing human suffering and misery. The scope of the suffering in Iraq, the scope of suffering that results from international combat is small compared to the suffering that you see resulting from car wrecks and the natural course of human life. We can create these technologies to address our our combat injuries, then we can alleviate so much misery in the world. I felt absolutely confident that a revolution in medicine was going to result from the conversations that were taking place at that event.
SUSAN DENTZER: So, taking for example Dr. Atala's work at Wake Forest on tissue engineering, as well as the prosthetics that you're able to devise, how do you see those coming together?
DAVID HANSON: It has to be an integrated systems approach. You have to surgically repair the face to accept the prosthesis. You have to grow the tissue to be compatible and embedded in the human living tissues or that has to converge with the surgery and the prosthesis has to be designed to fit into the face. There's also the possibility that you could use micro robotics embedded in the prosthesis or separate from a prosthesis as a smart bandage that actively cares for the wounded environment, sensing with a dense array of sensors and using micro fluid and network, a pore network similar to this sponge material.
SUSAN DENTZER: So how would it be possible to bring all these technologies together to help people like Jeffrey Mittman, or other wounded soldiers or other armed forces personnel?
DAVID HANSON:Every wound is different. So you would need a digital model of the soldier's wounds, digital models of the tissue solutions and surgical solutions as well as the robotic solution so that you could rapidly customize a solution for a given soldier's wounds. You'd have to coordinate your tissue engineers, your surgeons and the robotics teams to create a comprehensive solution.
There would be a custom prosthesis developed and tissues custom developed to restore the functionality of the face as much as possible. You would want to embed the tissues. You would design the tissues to be embedded surgically within the skin and you would design a prosthesis to fit into that. So you would need to be very aware of the boundary conditions between the prosthetic material and the living flesh. To maintain that balance between the artificial and the living, you would administer therapeutic agents as much as possible, emollients and the like to the living tissues underneath so you could have the prosthesis be active, smart, sensing the living conditions and applying therapies as well.
SUSAN DENTZER: So let's talk about Jules, the robot .
DAVID HANSON: This is Jules, who is a robot custom designed for the University of West England for cognitive science and artificial intelligence research. And Jules runs on very small batteries, like this one, a 9-volt battery. He is extremely portable and lightweight. He can run for about four hours on a laptop battery.
Jules makes a full range of facial expressions. For a conversational robot like Jules, you need all of the muscles represented, so that you can express joy, express anger, and sadness.
The face is packed with technology. You've got cameras in the eyes, you have active eyes that, that move, um, and you have uh, structures in the skin that imitate the facial tissues and the facial muscles. So motors that attach to lines that then are anchored into the skin. And the motors and lines and the way they move imitate natural muscle actions.
SUSAN DENTZER: So when he smiles what's happening?
DAVID HANSON: When he smiles, you have a synthetic zygomatic major muscle that pulls the corner of the mouth up towards the cheek. Then likewise when it frowns you have what's called the depressor muscle that moves it down. So you have a full array of natural muscles that are imitated in the face, then you also have sensors that can see human face, the human face. So for robotics that's very important, because you want to be able to make eye contact with people. You want the robots to engage people in a very natural way.
For a prosthesis you would also want an eye to make eye contact. So you have the gross motor skills in one side of the face being imitated on the other side of the face, but then you're going to need the thing to zero in so that it really locks eyes with you. So having a camera imbedded in the eye in an artificial eye will enable that eye contact to make a prosthesis uh more effective socially.
SUSAN DENTZER: You've done some research to better understand how people interact with robots, how they might approach a robotic or prosthetic face. Let's talk about that.
DAVID HANSON: We're very interested in how people socially interact with the robots. And we have been engaging in cognitive science, studies of how people react to artificial human faces. This is allowing us to tune the aesthetics more effectively. In the feedback we've received, most people found the robots to be very enjoyable to interact with and very few people found them to be disconcerting, but some people did. That's one thing that we have to overcome and learn more about.
We don't understand how we interact facially, how our facial expressions ground us in a conversation, what they mean. We understand that in a very superficial way. But the deeper semantics of facial expressions, what they mean socially, how they anchor us within our families and in our friends, is something that we have to address especially for restoring the cosmesis, or the cosmetics of a person's face who has suffered an injury.
SUSAN DENTZER: But at least your preliminary research would suggest that many people would not find a robotic, prosthetic face at all off putting. They'd look at a wounded person's face, perhaps bring to it a sense of empathy and sympathy because of what the person has suffered, be kind of charmed by the prosthetic aspect and, and be able to interact with it.
DAVID HANSON: Absolutely. We identify ourselves by our face. We identify each other by our, the way that your face looks and it's unconscious. We're highly evolved to perceive faces. And by restoring that face, we're removing the obstruction to socializing, to returning to a normal social interaction. The robotic aspect would be sensational but you would forget that you're interacting with somebody who has a partially robotic face very quickly, as you immerse yourself in the social conversation and restore your normal interaction.
SUSAN DENTZER: As you go back to being a human being communicating with another human being.
DAVID HANSON: Exactly.