Interview with Cai Bin, Founder of Unimagnetics, on the whole industry chain of quantum biomagnetic imaging in China
Quantum sensing, quantum computing and quantum communication are three important strategic directions in the field of quantum science and technology, among which, quantum magnetic field sensing technology has received key attention from various countries due to its wide application prospects and great commercial value in biomedical, engineering and geological, military and defense, and basic physics research fields.
In this context, in recent years, magnetoencephalography and magnetocardiography devices based on atomic magnetometer technology have started to enter the market. Atomic magnetometers not only cost less than SQUID, but also have the advantages of being sensitive to low frequency signals, room temperature operation, low power consumption, small size, wearable, etc. The detection sensitivity is close to or beyond that of SQUID, making them suitable for large-scale applications.
However, for many years the United States has monopolized the commercial technology in this field, and in April 2022, with the first domestic atomic magnetometer technology-based cardiac magnetograph to obtain a medical device registration certificate, China became the second country after the United States to have a full industrial chain of quantum biomagnetic field imaging, and to achieve this breakthrough is Beijing Unmagnetics Technology Co. ("Unmagnetics").
Recently, PhotonBox had an in-depth conversation with Cai Bin, the founder and chairman of UnMagnetic Technology, about the technology and application of quantum magnetometer.
Cai Bin, Founder and Chairman of the Board, UnMagnet Technology
The full text of the interview is as follows.
Why did you start Unmagnet Technology? What is Unmagnet Technology?
The original intention of Unimagnet Technology is to break the monopoly of foreign countries and break through the "neck" technology of commercialized quantum magnetic field sensor. After graduating from Beihang University, my study and work experience abroad for more than 10 years were basically in the field of aerospace and high-end manufacturing of special sensors, so the design of the underlying framework system of the quantum magnetic field sensor and many technical difficulties, including the equipment and processes related to mass production encountered in the industrialization process, in our eyes, are not too different from traditional high-end special sensors. The difference between the traditional high-end special sensor and the quantum magnetic field sensor is not too big. It is just that we can think of quantum magnetic field sensing is only the use of a new sensing principle. But after all, it is to become a commercial sensor product, whether it is our current medical devices or used for geological exploration, industrial applications, etc., it needs to meet certain common standards. Therefore, what I do is how to commercialize, miniaturize, and cost effectively a technical principle that is known to everyone in the world, to achieve the consistency and reliability required by the application scenario, and even to improve its technical index, to truly realize the transformation from lab to real world.
The founding of Unmagnetics was a coincidence, because the first project I started when I returned to China was a very special special sensor. Through this project, we learned about some very urgent needs of our country in the detection of very weak magnetic fields, and at the same time the gap with the US in atomic magnetometers was relatively large. At that time, it was 2019, the United States and the United Kingdom in biomagnetic field detection, such as the heart and brain magnetic results of some commercialization, so it was found that this is a very promising market direction, and at the same time, we can use our existing technical team in the relevant fields of more than ten years or even twenty to thirty years of experience, into a relatively more segmented, but high technical difficulty barriers, but also has a broad market prospects. track. In this context, we completed the establishment of our start-up team and have been working in this field since then along the lines of our goal back then.
Before the establishment of WMS, we knew very clearly that to make a breakthrough in this field, our technology team must have excellent scientists in the intersection of quantum physics, mathematics, optics, control, and materials science, as well as a team of engineers with strong experience in hardware productization and industrialization, in order to complete the process of advancing this technology from TRL3 to TRL9 (TRL - Technology Readiness Level).
The name "X-Magtech" has two main meanings: firstly, it refers to the future magnetic field related technology, and secondly, the word "un" also means unknown. At the same time, we borrowed the idea of SpaceX and X-Men, and the X stands for the 10 in Roman numerals, the largest number, and also represents the unknown and the future.
Our long-term plan is to make a great high-end special sensor company. In the near future, we are more concerned or want to become a company that can commercialize very advanced quantum technology into the lives of ordinary people as soon as possible, so our main research and development and promotion of the heart-brain magnetometer has become our grip to achieve our current goal.
How to realize from sensor to complete medical imaging device?
The problems faced in how to turn a quantum sensor into a system application are similar to those of traditional sensors, including its own collaborative work, the control part, including the environment it needs, etc., so its difficulties and challenges are very common. Since 2015, there are about 30 or 40 more famous teams around the world trying to buy the atomic magnetometer from Quspin in the US and then do the integration themselves. However, there are only a few commercialized products in the world, such as cardiac magnetics and brain magnetics, which were founded by the University of Nottingham, and Genetesis, which has completed several rounds of funding totaling $40M.
From here it is clear that the challenge of going from a small sensor to a complete commercial and especially medical system is huge. For example, we now have 36 channels in cardiac magnetism, and there may even be 72 channels in the future, such as brain magnetism system will have 32, 64, 82 and other configurations, so many high-density sensors will appear many academic papers mentioned the problem of crosstalk between each other, automatic cooperative control of large arrays. Of course, after commercialization, so many sensors working together in the case of consistency and reliability problems, which may be in the research community or in the functional prototype of some products do not need to consider the difficulties, but once it becomes a medical device, regulations on the reliability of the entire product will be very high requirements. To give you an example that you can understand, the principle prototype of our cardiac magnetometer was very mature about one and a half years ago, but it took us nearly a year to break through the reliability, stability and consistency of the entire array, and this was built on the background of our underlying atomic magnetometer technology. So it can also explain why so many people all over the world can buy atomic magnetometers. Without the in-depth technical support from magnetometer companies, it is difficult to achieve a commercially integrated cardio-magnetic system or we have not seen the results yet.
Besides, atomic magnetometers need a specific near-zero magnetic field environment when they work, on the one hand to provide a working background for the atomic magnetometer itself, and on the other hand because in this pure environment we can more effectively capture very weak magnetic fields like heart and brain and get a higher signal-to-noise ratio. This magnetic field shielding environment needs to be controlled at a certain cost, volume, weight and size, and the traditional solution is more of a service with scientific research applications, making a lot of concessions in these aspects, and there are generally too high prices, difficult to use, long construction cycle and other pain points, and now taking into account the needs of commercial practical application scenarios, we need to make a lot of innovation and optimization, and even break many conventions in order to find The balance between performance and practicality.
The last part is the software, in which the major difficulty and challenge is to achieve accurate examination and precise diagnosis for patients in the clinical setting, which is very different from the software and algorithm for cardiac and brain magnetic principle demonstration and research application.
After solving these technical difficulties, it is the traditional establishment of a full set of quality management system for the whole life cycle of medical device R&D, production and after-sales as well as related hardware facilities, but these steps are relatively easy to achieve. Overall, although I was just using cardiac magnetographs as an example, cardiac magnetographs or magnetoencephalographs are similar and face multifaceted and complex technical and commercial challenges.
Unmagnetic Technology Cardiac Magnetograph
What is the difference between the unmagnetized atomic magnetometer and the Quspin atomic magnetometer?
The main difference between our products and Quspin is the system architecture and software. Since we are targeting system-level applications, we have broken through many technical challenges from the beginning and implemented multi-channel in both hardware and software, while Quspin is currently shipping single-channel products and will launch multi-channel products in the future. This choice is mainly because we found in the early days that atomic magnetometers are basically large-scale array applications, and the proportion of customers using single channel is relatively small. In the biomagnetic imaging, which now has a clear commercialization prospect, the demand for a single channel will be even less. So one of the innovations we proposed when we did the product requirement analysis at that time was the direct implementation of multi-channel atomic magnetometer.
Multi-channel is mainly considered from two aspects, on the one hand, the integration of multiple channels is much better. For example, for a 64-channel magnetoencephalograph, in the case of Quspin's atomic magnetometer, it is necessary to integrate 64 complete atomic magnetometer systems and ancillary equipment such as hosts, power supply, signal transmission, and system management will be much more complicated. If we use our current 4-channel products, we only need 16 sets of products to solve the problem, so there will be a difference in magnitude, which becomes especially important in the pursuit of high reliability medical devices, while the system size and cost of multi-channel products will also have greater advantages.
Once the number and density of channels is high, the crosstalk between each other will become large or even uncontrollable, affecting the final imaging quality, while the multi-channel product architecture can be suppressed from the source of crosstalk.
What are the current products of Unimagnetics?
First of all, based on our core technology, Atomic Magnetometer, this product line includes multi-channel zero-field magnetometers that have been mass-produced and applied on the ground. atomic magnetometer, a full-field atomic magnetometer will certainly be introduced and commercialized.
Then there is the application level of atomic magnetometer, in the field of biomagnetic measurement, including the cardiomagnetometer, which has already received the medical device registration certificate, and the magnetoencephalograph, which is in the process of obtaining the medical machinery registration certificate. In these two product lines we will continue to upgrade and launch more specialized products targeting different target groups. For example, a higher performance cardiac magnetograph aimed at research users, and possibly a simplified portable product aimed at primary hospitals.
Zero-field atomic magnetometer from Unmagnetic Technologies
Who are the customers of Unmagnetics?
Our customers can be divided into two categories: the first one is the users who aim at scientific research, they mainly use the atomic magnetometer for a variety of underlying scientific applications, including biomagnetic field monitoring, but they may not simply measure the heart magnetism or brain magnetism, they may see more comprehensive, more detailed and more complete, such as muscle magnetism, cell magnetism and so on. They may also use atomic magnetometers for research instruments related to basic physics, or for the evaluation of very weak magnetic properties of special materials, or for the detection of very low electrical noise, all of which take advantage of the very extreme sensitivity of atomic magnetometers.
The second category is the customers of integrated products, mainly hospitals at present. These include our already licensed cardiac magnetographs and the magnetoencephalographs that are in the process of being licensed, both of which are very attractive to many hospitals and can be used in clinical and research applications according to demand.
But also includes customers targeting brain science brain cognition and brain-computer interface research.
What are the advantages of unmagnetized cardiac magnetographs and magnetoencephalographs compared to traditional testing instruments?
The current diagnostic techniques for ischemic and other heart diseases are mainly divided into electrocardiography and ultrasound, which are harmless to human body, and also include cardiac CTA, myocardial nuclide, coronary angiography, etc., which are damaging or radioactive to human body. Among the non-invasive techniques, cardiac ultrasound can only see the structural or blood flow abnormalities of our heart, lacking functional information; while electrocardiography, which responds to the functional aspects of the heart, can detect cardiac abnormalities often in the middle and late stages of the disease because the electrical signal is attenuated by the body and the detection sensitivity is limited. Invasive/radioactive tests require contrast injections for CTA, radioactive contrast injections for myocardial nuclide, and minimally invasive procedures for coronary angiography or FFR, in addition to radiation. In the early stage of the disease, non-invasive tests have low diagnostic accuracy and cannot be detected early; while invasive/radiated means with high diagnostic accuracy are not suitable for large-scale and frequent use, there is such a gap that cannot be accurately diagnosed in the early stage.
The biggest advantage of atomic magnetometer compared with other testing methods is that it can detect abnormalities in heart function at an early stage of ischemic heart disease without any damage to the human body, such as radiation, contrast, drugs or even surgery.
The reason for the absence of damage to the human body is that the magnetic field generated by the heart can penetrate the body without damage. An extreme example is that a fetus can have its heart wrapped in so many different body tissues of the mother, even including amniotic fluid and insulating fetal membranes, and still be detected by transmitting the cardiac magnetic signal to the surface of the body. When we have such high sensitivity and can detect the information non-invasively, the clinical value becomes very great.
In general, cardiac magnetography can help doctors to detect relevant diseases at an early stage, achieve early detection and early intervention, and ultimately reduce the mortality rate of serious diseases and serious illnesses. At present, our software has achieved real-time presentation, the scan time is 90 seconds, and the results are instantly visible after the scan is completed.
Can cardiac magnetography completely replace previous cardiac testing methods? Do physicians who use cardiac magnetographs need to learn how to use the equipment?
No. Cardiac magnetography is an important tool for cardiac function testing, providing clinical experts with more comprehensive information about the state of the heart and helping them to choose the most appropriate diagnostic device based on the type of disease or patient characteristics, so it is complementary to existing tools.
How to let doctors or technicians learn to use the cardiac magnetograph equipment and analyze the data efficiently is the goal we always consider when developing the cardiac magnetograph, through our efforts, we have now achieved to fully automate and intellectualize the complex operation and data analysis of the cardiac magnetograph. The training time is only about one hour and includes the whole operation process of hardware and software.
In addition to the atomic magnetometer, what are the key technologies that need to be overcome in the development of the magnetoencephalograph?
Magnetoencephalography is very similar to cardiac magnetography at the system level, but the first difference is in the sensitivity of the system. The second is in software data processing.
Driven by these two differences, our magnetoencephalograph products must first achieve a higher level of atomic magnetometer and magnetic field shielding than the cardiac magnetometer, but of course the number of atomic magnetometers and the way they are arranged also pose some challenges. For example, the heat generation, the reduction of crosstalk in the case of higher sensor density and closer distance, the high reliability operation of complex systems, etc.
Secondly, the data processing in the data software, on the one hand, is the raw data pre-processing related algorithms, which need to be developed according to some characteristics of our sensors and special algorithms. On the other hand is the traceability and imaging we mentioned earlier, but this part is more of a software implementation since it is not very different from the traditional superconducting quantum interferometer, and there are methods that have been used for many years and are widely accepted.
Besides the medical field, what are the other application scenarios for atomic magnetometers?
As I mentioned earlier, atomic magnetometers include zero-field magnetometers (QZFM) and full-field magnetometers (QTFM).
From the state of the art, if you want to apply the magnetometer on the application scenario of very weak magnetic field detection of biomagnetic field, you must have miniaturization and high sensitivity, and also need relatively controllable cost as well as reliability, relying mainly on the zero-field atomic magnetometer. The full-field atomic magnetometer is aimed at a completely different application scenario, mainly used in the detection of weak magnetic fields under the geomagnetic field, such as geological exploration, non-destructive testing and military defense, etc. The industry has been using Quspin's products to explore a variety of applications, it is worth mentioning that because the full-field magnetometer can be used in military defense, so Quspin's related high-end products are embargoed to China It is worth mentioning that Quspin's high-end products are embargoed to China, so this becomes our next challenge to the "neck" problem.
What did Quspin do right when it completed four rounds of financing within two years?
The frequency and strategy of our four rounds of financing is more in line with the current hard tech startups, starting from tens of millions in the angel round, then rising to tens of millions in the PreA round, and then to over a billion in the A round. What our investors see as our shining points should be three things.
The first point is the track. Of course, the track is not the same for different investors, some investors are interested in the whole quantum sensing track, some investors may value the medical device track. No matter which track these two tracks are, it is full of imagination and is a track with large enough market space.
The second point that many investors may value is the value of the core technology that we have mastered. Including the principle of atomic magnetometer, production, research and development, mass production, to the system integration capabilities, and even to our current continuous iteration of continuous research and development capabilities are very valuable.
The third point is the team. Especially in the early days, most investors still value the reasonable division of labor and rich experience of the team, as well as the more mature team internal friction. After more than two years of validation, we are already a team with recognized execution and fighting ability.
What is the next step in the development of Unmagnet?
Our next development plan is to commercialize and promote our relatively mature products such as Atomic Magnetometer and Cardiac Magnetometer in the market; in R&D, we will continue to upgrade and expand our existing product pipeline, and develop more pipelines and enter into different industries.
