OFweekRobotOn July 7, the 2nd CCF-GAIR Global Artificial Intelligence and Robotics Summit, hosted by the China Computer Federation (CCF) and hosted by Lei Feng.com and the Chinese University of Hong Kong (Shenzhen), opened as scheduled in Shenzhen. In the medical artificial intelligence session on the third day of the conference, Xi Ning, chair professor of the University of Hong Kong, director of the Institute of Advanced Technology, and chairman-designate of IEEE RAS, delivered a conference report.
Professor Xi Ning believes that human’s understanding of life has reached the scale of molecules and cells, and the innovation of drug development and treatment methods requires a method of measuring and operating on the scale of molecules and cells – micro-nano robots. He further cited several examples of micro-nano robots used in new drug development and diagnosis and treatment. He believes that micro-nano robots expand human capabilities and allow people to explore nature to the nanoscale, which will bring subversive effects to all aspects of human beings.
The following content is organized by Lei Feng.com from Professor Xi Ning’s report, with deletions:
Today I would like to introduce the application of micro-nano robots in medicine and new drug development.
We know that medical treatment is mainly about diagnosis and treatment. From an engineering point of view, diagnosis is the measurement of various abnormal phenomena in the human body. Treatment, in terms of engineering, is to change the existing state of human cells. The two important functions of robots: Sensing and Manipulation can play an important role in diagnosis and treatment.
We first review the history of human beings’ struggle against diseases. At the beginning, the diagnosis was based on whether the patient’s complexion was good, and Chinese medicine looked at the tongue coating, complexion, and complexion. With X-ray technology in the future, you can see the inside of the body, see the lungs and other organs, and diagnose the patient’s physical state from the shape of the organs. Now that medicine has developed to the cellular and molecular stage, the same is true of medical diagnosis. We need to reduce the diagnostic technology to the scale of microns and nanometers, and perform Sensing at the scale of cells and molecules to provide new diagnostic methods.
The same is true for treatment. It starts from the perspective of organs, and operates on organs. Taking Western medicine as an example, the repair and removal of organs are all on this scale. With the development of molecular biology, surgery has begun to be carried out at the cellular level and at the level of DNA molecules, while DNA molecules are the core code that controls the growth and development of cells, and cells are the basic units that make up organs and tissues, which will be solved from more fundamental reasons. human disease problems. At the same time, as the scale decreases, Sensing and Manipulation methods on smaller scales are required.
Why do new drug development need this?
The most traditional Chinese medicine is summed up by human beings through hundreds of thousands of years of accumulation and continuous experiments, and it is the result of time accumulation. The traditional new drug development is based on the target-based new drug development, that is, the new drug development based on the signaling pathway, which comprehensively sees the relationship between the disease and various factors, and considers the side effects comprehensively. The development of new drugs requires new means and new technologies, namely Sensing and Manipulation at the molecular and cellular levels, which also provides new application areas for the development of robots.
Drug development is an important part of humanity’s struggle against disease. Humans are now facing a great challenge. The cost of new drug development is getting higher and higher. It takes 1 billion to 1.5 billion US dollars to develop a drug, which takes nearly 10 years. At present, the investment in new drug development is increasing every year, but the number of drugs is basically the same, indicating that there are few new drugs, and the gap between input and output is increasing.
At the same time, there are more and more new diseases. Many diseases were unheard of decades ago, but now there are many new diseases, the number of drugs has not increased, and the cost is getting higher and higher. This is a big problem for human beings in the fight against disease. Solving this problem requires opening up new avenues for new drug development, and new technologies must be harnessed to change the status quo.
Robot andautomationCan technology help us solve this problem?
Artificial intelligence and robotics are hot fields, playing a big role in the manufacturing industry and life, and also in new drugs and medical diagnosis and treatment.
Robots were originally used to replace people (machines to replace people) to do things that people can do but are unwilling to do, such as highly repetitive labor, but with the development of robotics, robots have changed from simply replacing people to expanding. People are not only doing things that people don’t want to do, but also doing things that people can’t do. For example, the combination of robots and modern information technology and network technology enables robots to operate in long distances and help people overcome the difficulties caused by distance.
1 2 3 Next > page
The Robot operates and measures in a very small scale, which means that it can overcome the difficulties brought by scale to humans, and operate and measure in an environment where people can’t see or touch. In this way, it has many applications in medical diagnosis and new drug development. At the same time, the robot can also enter the environment that is difficult for people to enter, such as the physiological environment, enter the human body, and help people overcome the difficulties caused by the environment. The scale is too small, the distance is too special, and the environment is too special for humans to do, but with the help of robots, we can overcome these difficulties. Robots help us go beyond the limits of human capabilities, beyond the limits of scale, distance, and environment.
This is the ultra-limited robot, which allows robots to expand the limits of human beings and do things that humans cannot do. This technology has great applications in new drug development and medical diagnosis.
In the traditional concept, robots play a great role in manufacturing, such as automobile manufacturing. We are now thinking about how to transplant the technology of automated robots in automobile manufacturing and manufacturing to new drug development, medical diagnosis and treatment, and to overcome the challenges faced by human beings in new drug development and medical diagnosis that I just mentioned.
The potential economic value is very large.According to statistics, the carindustryThe output value of 728 billion US dollars, of which 508 billion is the value brought by the use of automation and robots. In other words, without robots and automation, there would be no automobile industry today, and there would be no material civilization that humans have today.
The global pharmaceutical industry is larger than the automobile industry, and the degree of automation in the development of new drugs is very low. Robots have been used on a large scale in automobile manufacturing, but most of the research and development of new drugs is done manually in the laboratory, and the degree of automation is very low. According to statistical principles similar to economics, if we use robotics and automation for new drug development, the potential economic value is enormous and far exceeds the automotive industry.
Nanorobots for new drug development
How can we apply the concepts of automation and robotics to new drug development?
We imagine that, like today’s assembly line, the conveyor belt continuously feeds the cells into the working space of the robot, and the robot puts different drugs on the cells and measures the cells at the same time, measuring different targets, measuring the effect of different drugs .
The automation of the whole process has several key technologies to achieve this goal: one is automated transportation; the other is automated drug delivery and automated measurement of drug delivery.
This is basically achieved in both the automotive industry and the manufacturing industry. But the biggest difference is that the assembly of the car, where all the parts are about the same size, is a structured environment where people make everything. However, in the development of new drugs, each cell grows differently and is unstructured, so it poses new challenges to robotics in terms of sensory control and planning.
In order to meet the challenge, we have done work in several areas. First of all, the robot must have the means of operation. We have developed a nano-manipulation robot, which can operate and measure objects at the nano-scale, which is very important. For robots to operate, the most important thing is to make the invisible and intangible things both visible and touchable, so that the medicine can be placed in a specific position and the effect of the medicine can be measured.
For example, the one shown on the right in the above image is a nanomanipulator, which is operated by a human with a joystick. The robot can move in it. Below the picture on the right is a cell. We can do some operations, remove some parts or put some drugs on it.
To do this, the nanoscale environment needs to be displayed so that people can see it and operate it. We’ve discovered a technique that enables high-speed imaging at the nanoscale, producing real-time video-like images that can aid in manipulation. We use the principle of compressed sensing, because the scale is very small, we need to measure the environment at high speed, because the optics are invisible, and the scale is too small, we need to use atomic force microscopes and electron microscopes to measure.
The image above is a real-time image of DNA molecules sloshing around in a liquid. We can measure the movement of DNA molecules in real time, which is very small scale because DNA molecules are only 1-2 nanometers in diameter.
This is the basic stuff to manipulate and measure. With this in place, we can proceed.
On the left is a DNA molecule, and the red one is a nanorobot. We let the nanorobot move along the DNA molecule and keep it on the DNA molecule, so the precision of position control is required to be 1-2 nanometers. What it does is, we know that the most important thing about DNA is sequencing, and that sequencing can predict future diseases. Sequencing is the measurement of four different molecules of ATCG, and if a nanorobot walks along, it can quickly know the molecular composition, which is the way to quickly sequence DNA.
Imaging is important, but at the same time operating. The operation needs to be performed on the micro-nano scale. The operation requires the end effector of the robot to have a certain rigidity, while the imaging perception requires the end effector to be very soft, so there is a contradiction between operation and imaging. For example, if a nano-scale probe is small and soft, when you try to push something, it will become bent and unable to push, which brings great difficulties to the operation at the nano-scale. However, the stiffness cannot be made very large. If the stiffness is too large, if you touch a very soft object, you will not know its softness and hardness, which will reduce the accuracy of sensing and measurement.
Our solution to this contradiction is to add a driver to the probe, and the probe can be changed by controlling the driver.mechanicalThe characteristics make the stiffness of the probe adjustable, and it becomes very soft when measuring, and it is easy to measure the hardness of the environment. At the same time it becomes very hard when operating.
For example, the nanowire in the picture above is only about 100 nanometers, which is one thousandth of the diameter of a hair. The black one is the probe. After we add the control signal, the probe itself becomes hard and can be pushed. The probe can achieve our purpose by changing the mechanical properties. This is a very important example of micro-nano manipulation.
Nanorobots for diagnosis and treatment
Robot-related technologies, what are the applications in medical diagnosis and treatment?
Such as the treatment of psoriasis in skin diseases. Psoriasis is an immune disease. There is a protein Desmosome between epithelial cells and cells in human skin. Due to some immune diseases, the body produces an antibody (a protein) that attacks and destroys the Desmosome, forming a lot of blisters on the surface of the body, and it rots. At that time, the mechanism of the whole process was not clear. Some people speculated that it was the cause of the antibody, and some people speculated that it might be caused by the process of signal transduction.
It is very difficult to study because of environmental limitations. Our approach is to mechanically cut the Desmosome apart with nanorobots. Or use nanorobots to put the antibody directly on the Desmosome and see if it will be destroyed. By comparing the effects of mechanical cleavage and antibodies, the similarities and differences between the two can be tested to help determine the cause of the disease.
Finally, our research proves that this skin disease is not caused by mechanical action, but caused by signal transduction. Therefore, the improvement of tools and technologies has made it easier to study problems that were previously difficult to study.
We can also use this method to study the breakdown of stem cells. Stem cells are very important, but it is very difficult to predict when and under what conditions they decompose, and to measure the state of decomposition. Thanks to nanorobots, the mechanical properties of some stem cells can be measured in real time to predict the state of decomposition.
Another example of successful application of nanorobots in the field of treatment is lymphoma. There is a specific drug for lymphoma treatment, Rituxan, which is a targeted drug and has been used with great success in clinical practice. But there are differences in drug resistance, that is, the drug works for some people and not for others. The drug is expensive and the treatment cost is high. If the treatment effect cannot be known in advance, it will not only waste money but also delay the precious treatment opportunity. So there is a need for a way to predict how well the treatment will work before treatment.
We used nanorobots to remove cancer cells from patients, and found that only when the cancer cell target and drug binding force reached a certain level, it would work. Through this study, it is possible to predict what the effect of targeted therapy will be, which is of great clinical significance.
Another example is the study of cell adhesion. The size of cell adhesion directly affects wound healing. It also has a very important role in the field of prosthetics. Now there is a successful method, which is to insert steel pipes into the bones, so that the effect of the prosthesis is the same as that of the real person. But there is a big problem inside, that is, the steel pipe is inserted into the leg, and the skin and flesh must grow outside the steel pipe, but there are often gaps, which will cause bacteria to cause infection. After a long time, it will cause bone infection, which will eventually have to be removed. prosthesis. People expect to study the adhesion of cells, especially the adhesion between cells and prostheses. After understanding the mechanism, a series of methods can be used to make it adhere well and prevent infection.
It’s hard. For example, how to measure cell adhesion? A method has been developed through nanorobots that can be measured in real time to study the effect of different drugs on adhesion.
Another application is the measurement of ion channel ionic currents. Real-time measurement of the current of ion channels is of great significance for understanding the physiological functions of cells and treating many diseases. But the measurement is difficult. The traditional method in the past is called patch clamp, which is a kind of technical work, and it takes many years of practice to measure. Now, nano-robot technology can be used to locate and measure accurately, which makes the original complicated process very simple, and can measure at high speed.
There are inner ear cells in the human ear, and there are many cilia on the surface of the inner ear cells, just like antennae. When the vibrations in the air are transmitted to the cochlea, it causes the cilia to bend and deform, which in turn opens specific ion channels, generating signals and making people hear sounds. Many people are deaf due to problems with ion channels. Therefore, it is necessary to study a drug to change this phenomenon, but in the process of studying the drug, there must be a way to measure it in real time to see if the drug can make the ion channel normal.
However, this kind of measurement is difficult, because it is necessary to measure the ion channel on the cell, and at the same time generate mechanical stimulation to make the cilia bend and deform, which requires precise manipulation on a very small scale, which is very difficult. Using nanorobots, we can not only perform ultra-fine and ultra-precise mechanical stimulation of these cilia, but also measure ion channels, so that we can try different drugs to understand their therapeutic effects.
To sum up, human’s cognition of life has reached the molecular and cellular scale. Therefore, whether it is drug development or innovation of therapeutic methods, tools for measurement and manipulation at the molecular and cellular scales are required, thereby reducing traditional operations from the level of organs to the level of molecules and cells, thus helping us to carry out medical diagnosis, Help us in the development of new drugs, resulting in many new methods of diagnosis and treatment to deal with the diseases facing human beings.
The Links: 3HAC020466-001 JZRCR-XSU02 DISPLAY