Feynman's Rhapsody: The 'surgeon' who can enter the body

In 1959, Nobel Prize winner in physics Richard Feynman gave a lecture at the California Institute of Technology entitled "There is Plenty of Room at the Bottom" (There is still a vast space for bottom-level research) ’s speech first envisioned the possibility of microrobots inside the body.

In Feynman’s conjecture, this type of micro-robot is driven by Micro-Electro-Mechanical System (MEMS) and can enter the body to perform surgery. At that time, Feynman said: "If we can swallow a surgeon, then many complex operations can become very interesting and simple."

Perhaps Feynman's influence is too great. In ten years Later, before the scientific research community could conduct research, American director Richard Fleischer made Feynman's idea into a classic science fiction movie, "Fantastic Voyage." In the movie, five doctors were reduced to one millionth of their original size and injected into the body of a life-threatening patient whose cerebral blood vessels were damaged. After a series of adventures, they finally succeeded in finding the bleeding point and saving him in time. took the life of the patient.

But, can micro-robots that can enter the human body only be a fantasy? The answer is obviously doubtful.

Since Feynman proposed the concept of "in vivo surgeon", scientists have been attracted and inspired by this idea, invested in the research of microrobots, and achieved many good results. Scientists imagine that in the future, machines can really enter the human body to achieve targeted therapy and drug delivery, and help treat major diseases such as tumors.

1. "Slime" Robot

Some time ago, a magnetically controlled slime micro-robot called "Slime" became popular on New Scientist.

It is made of magnetic mucus material and can be entered into the body to remove small devices accidentally swallowed. As soon as it was released on April 1, it immediately caused a huge sensation in the technology community. Netizens were shocked and the click rate quickly exceeded 100,000, 1,000,000, and 10,000,000:

## Different from our common robots, this robot is more like a "monster" in terms of appearance, movements and abilities. It is very different from the "machine" and "human" that we imagine are made of rigid hardware and have facial features and bodies that resemble humans. In and out.

According to the Demo display, its appearance is like a lump of black dough, with no head, face, hands or feet. It has a soft body and various shapes.

But despite its ugly appearance, this "slime" robot has a variety of unique functions. It can transform its soft body, pass through narrow gaps, repair broken wires, and can also move inside the human digestive tract. Remove accidentally eaten electronic components by swallowing them.

Even if it is cut into several pieces and then spliced ​​back together, it still has the ability to "self-heal".

The "Slime" robot has broken most people's traditional impressions of "robots", and with its cool futuristic feel and potential power, "microrobots inside the body" have entered the public's field of vision.

One of the developers of the "Slime" robot, Professor Zhang Li of the Chinese University of Hong Kong, said that "self-healing" ability is also one of the current popular research directions in the field of soft robots, mainly reflected in the ability to respond to different environments. High adaptability. The power of the "Slime" robot is not only that it can heal itself after being cut off, but also that it can maintain its complete shape even if it is placed in a liquid, and can even travel unimpeded in air and solid environments. .

Illustration: Zhang Li, Professor of the Department of Mechanical and Automation Engineering of the Chinese University of Hong Kong

In fact, in recent years, research results on micro-robots have emerged one after another, such as A flying robot the size of a fly/beetle driven by micro motors. Different from them, the unique feature of the "Slime" robot is that magnetic powder and magnetic particles are added to the non-Newtonian liquid material to achieve magnetic field control, making it flexible and even able to deform freely.

The "self-healing" ability displayed by the "Slime" robot also relies on the characteristics of the material itself, coupled with the magnetic guidance of the outside world. Objects interact with each other and become one again.

It should be noted that the "Slime" robot currently does not have the ability to move autonomously. Its movement and expansion rely on the external magnetic field to control the neodymium magnet inside (which can be understood as a "powerful small magnetic ball").

"The 'Slime' robot itself has no fixed shape. It is slime-like. When a magnetic field is added, it will respond to the magnetic field. If you move the magnet from left to right, it will follow Hold the magnet from left to right. Due to the size of the magnetic force, it can easily change its shape. For example, if someone accidentally swallows a harmful component, turn it into a hand and grab the package." Zhang Lixiang commented on AI Technology introduce.

This is also the first time that the Advanced Nanomaterials & Microrobotics Laboratory (ANML) led by Zhang Li has produced a magnetic slime robot. Previously, ANML has produced many different types of micro nanorobots, including bionic insect robots printed with 3D technology, all of which are remotely controlled based on magnetic fields. "But slime-like robots like the 'Slime' robot have such large deformations." , it can be rolled up like an elephant's trunk, this is the first time." Zhang Li said.

What’s even more amazing is that it only took half a year for Sun Mengmeng, the first author of the work and an in-service postdoctoral fellow in the ANML laboratory, to join ANML, start research and development, and publish the article.

Illustration: Dr. Sun Mengmeng

"This is mainly because Sun Mengmeng already had some knowledge when he was a PhD student at Harbin Institute of Technology (under the tutelage of Professor Xie Hui). Idea. After coming here, with the help of our research group’s extensive experience in related materials and magnetic control operations, the project progressed smoothly." Zhang Li introduced.

Considering the complexity of the internal environment of the human body, Zhang Li’s team imagined that the “Slime” robot might have a certain application space in the digestive tract. There are several main reasons: First, the cavity of the digestive tract If it is larger, the "Slime" robot will shuttle more smoothly inside; secondly, the human digestive tract already has many microbial flora, so the risk of trying a robot inside the body is relatively low; thirdly, the materials used to make the "Slime" robot After cytotoxicity testing, the toxicity is low. If it only stays in the body for a short time and is excreted, it is theoretically safe.

Of course, the idea of ​​using "slime" robots as internal treatment actuators is still at the conceptual stage and needs further exploration.

2. Development of micro-robots inside the body

The performance of "slime" robots is gratifying, but looking back at the development history of such micro-robots that can enter the body, it only lasts a few decades.

In the 1970s, in order to promote classified research, the U.S. intelligence agency tried to design some micro-robots that could perform prisoner-of-war assistance and electronic interception tasks. However, because the underlying supporting technology was not fully developed at the time, the micro-robot prototypes were not developed. Nothing was developed from this early set of calculations and concepts.

It was not until the 21st century that microrobots were officially launched. With the development of multidisciplinary fields such as microelectromechanical and microactuators, microrobots have achieved important technological breakthroughs and gradually become an international research hotspot.

Illustration: Bionic Micro Robot

Compared with large robots that have been studied for more than half a century, the development of micro robots only lasts for more than 20 years. There are only a handful of "microrobots that can enter the body", and they are in their infancy both at home and abroad.

There are different categories of micro robots. Among them, micro medical robots are considered by the industry to be the most promising application field. The Japan Science and Technology Policy Research Institute has predicted that “in the future, surgeries using microrobots and robots in the medical field will account for more than half of all medical surgeries.”

Abroad, Japan has taken the lead in adopting the "robotic surgeon" plan and is developing ultra-miniature robots that can travel through human blood vessels to find and kill cancer cells. The John Hopkins Laboratory in Maryland, USA, has developed a miniature detection device equipped with a miniature silicon thermometer and a miniature circuit. When swallowed into the body, the body's temperature information can be sent to a recorder. Swedish scientists have invented a robot that is as big as an English punctuation mark. In the future, it can move single cells or capture bacteria to perform various surgeries in the human body.

Domestic researchers have also paid attention to this cutting-edge direction early, such as Professor Sun Lining of Suzhou University and Professor Liu Lianqing of Shenyang Automation Research Institute. In the field of "in vivo robots", young scholars such as Zhang Li, a professor at the Chinese University of Hong Kong, and Xu Tiantian, a researcher at the Shenzhen Institute of Advanced Technology, are not far behind, exploring new opportunities from the two major directions of materials and control.

In general, there are three major elements for the realization of robots in the body: first, the realization of a "micro" body shape; second, safety materials that adapt to the internal environment; third, the "automatic operation of the robot inside the body" driving skills.

Take the "Slime" robot as an example. Its biggest breakthrough is the material. It uses polyvinyl alcohol and borax materials with non-Newtonian fluid characteristics, plus a layer of silica. Its viscosity changes with contact with the outside world, and it has high adaptability to the environment. It can be used in gaseous, liquid and solid environments. Both can extend and crawl, and can perform multi-modal manipulation.

Illustration: Pour non-Newtonian liquid into the pool to float on the water

However, the toxicity safety of borax has not yet been guaranteed, and this field is currently A focus of research is to identify materials more suitable for building tiny medical robots. The material must be flexible, skin-friendly, non-toxic, harmless, easy to eliminate from the body, and easy to operate.

Regarding innovation and safety, Professor Zhang Li’s view is: “Sometimes scientists and doctors have different ideas. Doctors are often more conservative and often consider safety first, while scientists emphasize innovation more. There is a certain contradiction between the two." But in medical scenarios, there is no doubt that safety must be the first priority.

In addition to materials, controlling the path of microrobots in the body is another problem that needs to be solved urgently to become a "surgeon". In recent years, the research focus of in vivo microrobots has experienced three stages of changes: from open-loop control to closed-loop control, from a single motion mode to multiple motion modes, and from a single robot to multiple robots. The control of micro-robot clusters has practical application value in in-vivo medical scenarios and is also a major research trend in the field of robotics.

Compared with a single robot, cluster microrobots have two major advantages:

First, reduce the failure rate. For example, for drug delivery, the drug loading dose of swarm robots can be increased. In addition, in environments such as blood, a single tiny robot can easily be washed away by blood or engulfed by macrophages. At this time, switching to a swarm of robots can improve the success rate of treatment;

The second is swarming Easy to observe. Today's robots can reach nanometer scales, but when they are placed inside the body, it is extremely difficult to clearly observe a single robot using existing medical imaging equipment. Just like diving, we tend to ignore a small fish swimming in front of us, but we are often shocked by a group of dark fish in the distance.

3. Path control: "driving" in the body

In terms of path control of micro-robots, Xu Tiantian, a researcher at the Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences (referred to as "Shenzhen Institute of Advanced Technology") is an expert. A “new star” in scientific research.

Xu Tiantian comes from a background in automation control. He received his master's and doctorate degrees from the Ecole Centrale de Paris and the University of Paris VI. He began researching micro-robots during his Ph.D. After graduating with her PhD in 2014, she joined Professor Zhang Li's team at the Chinese University of Hong Kong as a postdoctoral researcher. In 2016, she officially joined the Intelligent Bionic Center of the Institute of Integration, Shenzhen Advanced Institute of Technology. She is currently the only scientist at Shenzhen Advanced Institute of Technology who studies the path control of medical microrobots.

Illustration: Researcher Xu Tiantian, Shenzhen Institute of Advanced Technology

From Professor Xu Tiantian’s perspective, research on path control of microrobots in vivo can be roughly divided into three directions. : First, how to make micro-robots move inside the body? The second is how to make them move according to the established path? Third, how to adapt them to the complex environment in the body?

If microrobots are compared to cars, then the suspended movement of the robot inside the body is equivalent to controlling a car to drive in the air in a complex and busy city. It is extremely difficult and has extremely high safety risk factors.

It should be noted that many physical laws in the microscopic world are different from those in the macroscopic world. For example, in 1976, Nobel physicist E.M. Purcell proposed the "scallop theorem", which means that when a scallop opens its shell quickly and then slowly closes it, due to inertia, the scallop will jump forward when it is opened quickly. It forms a "movement forward" that runs one after another. However, in the microscopic world, since the inertial force is almost negligible in the face of viscosity, the opening and closing actions of the scallop cannot move it forward.

#The internal environment of the human body is also a microscopic world. How to make microrobots move inside the body?

Xu Tiantian worked with the team to draw inspiration from nature: one is E. coli, which is driven forward by a spiral tail, just like twisting a screw, turning and moving forward; the other is flexible sperm Vibrate, vibrate forward by flapping its tail. In these two ways, they successfully created spiral-shaped robots and sperm-shaped bionic robots, and successfully made the robots move in an environment that mimics the environment inside the body.

Illustration: The micro-robot moves forward by "spiral swimming" in the liquid

However, it is not enough to make the robot move inside the body. It is necessary to ensure that the forward path is safe and cannot run rampant inside the body...

Therefore, in order to ensure that the robot shuttles "accurately" in the body, bypasses dangerous areas, and ensures safety, the path control of micro-robots is studied appears particularly important. And as mentioned before, the robot operates in the body by "moving forward in the air," which requires the robot to have 3D movement capabilities.

In 2019, Xu Tiantian’s team proposed a new type of path following control algorithm, which uses the path differentiation method to differentiate any given path into various small segments, allowing it to find its nearest small segment at each point. , to control its forward direction. Their algorithm successfully realized 3D path control for millimeter-level magnetically driven soft robots. The related work won the IEEE International Conference on Intelligent Robots and Systems (IROS) Best Application Paper Award:

In terms of path control, Xu Tiantian’s team also adopts magnetic control. The main advantage of magnetic control is that it can be controlled wirelessly: if the robot enters the body, a human researcher or doctor can perform the operation outside the body. At the same time, the magnetic control has a short response time, high power density, and high repeatability. The robot can successfully reach the lesion multiple times, eliminating the randomness of the success rate.

Note: Xu Tiantian’s team’s multi-degree-of-freedom magnetic control device

After realizing the 3D path control of a single robot, Xu Tiantian and his team turned to multi-robot control. Collaborative control research advances.

Xu Tiantian explained to AI Technology Review that there are two major difficulties in the operation of micro-robots based on magnetic control: First, the input signals in the same magnetic field are the same, which will lead to the direction of movement of multiple micro-robots. It is consistent with the speed; second, there is a lack of communication between micro-robots and they cannot be controlled independently.

In order to solve this problem, Xu Tiantian and his team have studied for many years, and finally achieved results at the beginning of this year-

They proposed a completely decoupled method that does not require communication and uses external unified signals. To perceive the robot and solve the problem of how to produce different outputs for the same signal, it achieved the independent position control of four magnetic soft micro-robots and the independent path following control of three magnetic soft micro-robots for the first time. Related work ("Independent Control Strategy" of Multiple Magnetic Flexible Millirobots for Position Control and Path Following” was published in T-RO, the top international robotics journal.

Legend: Independent position control of millimeter-level robots: (a) position control of two robots; (c) position control of three robots; (e) position control of four robots; (b), (d) and (f) are the corresponding position trajectories of the robot

This work is a big step forward in the collaborative control of multiple micro-robots. However, Xu Tiantian also told AI Technology Review that currently they have only achieved independent control of four micro-robots, and in the future, they will move towards greater goals.

It is worth noting that the introduction of artificial intelligence algorithms into path control is also becoming a trend. For example, Xu Tiantian and others began to use the "width learning" method proposed by Chen Junlong, Dean of the School of Computer Science at South China University of Technology in 2020, to automatically calculate and optimize the control rate of robots in complex environments, thereby achieving better control. .

4. Imagination and reality

So, how long will it take before micro-robots enter the body?

There is no doubt that Feynman’s conjecture is very avant-garde, and the idea of ​​an “internal surgeon” is also very fascinating.

Some time ago, Nature also published an article discussing the prospects of microrobots for cancer treatment. For example, anticancer drugs often take a shotgun approach, with traditional treatments involving intravenous injections of clotting drugs, which come with the risk of blood clots. While chemotherapy destroys tumors, it inevitably attacks healthy cells, causing a series of side effects. The coveted alternative to this dilemma is to inject a microrobot into people with cancer for targeted therapy and drug delivery.

Imagining that micro-robots can one day enter the body for cancer treatment, Zhang Li has great research enthusiasm and motivation. But at the same time, researchers are also clearly aware that there is still a long way to go before the implementation of microrobots in the body. For example, so far, no researchers at home or abroad have actually implemented microrobots in the body. Leifeng.com

Safety, ethics, cost performance, risk control, etc. are all issues that people will have to solve in the future.

Scientists are working hard to promote the research and implementation of internal robots. Zhang Li told AI Technology Review that in recent years, the Hong Kong government has invested HK$470 million to build a medical robot innovation technology center in the Hong Kong Science Park (pictured below), which is equipped with technologically advanced medical imaging equipment, magnetic resonance technology and X-ray technology. etc., to help scientists carry out innovation and technology incubation of medical robots.

The picture is provided by Professor Zhang Li

"From a scientific research perspective, I don’t think the 'Slime' robot is a landmark Innovation." Zhang Li said, "What we hope to achieve more is to give micro-robots intelligence, make breakthroughs in micro-robot clusters and control systems, make devices safer, smaller, and more intelligent, and then find it The ultimate goal of exporting medical applications is to benefit mankind."

Perhaps the idea of ​​an "internal surgeon" proposed by Feynman in the 1950s will be realized in the near future. , in the future it can be applied to any part of the human body, such as the fundus, retina, gastrointestinal tract, bladder or blood vessels.

Let us look forward to this day coming soon.

Reference link:

https://www.nature.com/articles/d41586-022-00859-0

https://twitter.com/newscientist/ status/1509599345255100417

https://www.siat.ac.cn/yjdw2016/rcdt2016/201912/t20191206_5449581.html

https://en.wikipedia.org/wiki/Microbotics

https://cuhk.edu.hk/chinese/features/zhang_li.html

http://www.cuhklizhanggroup.com/

http://people. ucas.edu.cn/~xutiantian

https://m.xzbu.com/9/view-9606955.htm​

The above is the detailed content of Feynman's Rhapsody: The 'surgeon' who can enter the body. For more information, please follow other related articles on the PHP Chinese website!