Quantum computing has stepped out of "science fiction movies". Boson Quantum has delivered over a hundred real-world application cases in fields such as oncology and brain-computer interfaces.
Text by | Hu Xiangyun
Edited by | Hai Ruojing
In the science - fiction movie "The Wandering Earth 2", the intelligent quantum computer "MOSS" with exponential computing power can not only coordinate the world's computing resources to support the synchronous operation of 10,000 "planetary engines", but also meet the massive and almost bottomless computing needs of the "Digital Life Project", arousing the public's infinite imagination about quantum computers. This once amazing science - fiction setting has now gradually moved from the laboratory to commercial implementation.
According to the "15th Five - Year Plan Outline", quantum technology has been listed as one of the six future industries. The market is also very enthusiastic. According to IT Juzi data, by April this year, there were more than 90 enterprises in the domestic quantum computing track, and the total valuation of the top 10 enterprises was approaching 50 billion yuan. Among them, market - oriented teams with top overseas academic backgrounds and mature industrial experiences are more favored in this round of capital competition, and their valuation levels are relatively higher than those of similar enterprises.
Bose Quantum, founded in 2020, is one of them. Although Bose Quantum is not attached to any scientific research institute, it has received 11 rounds of financing, with a cumulative amount of 1.84 billion yuan. The 1 - billion - yuan Series B financing completed in March this year directly set a record for single - round financing in the domestic quantum computing field. In terms of team composition, Dr. Wen Kai, the founder and CEO, graduated from Stanford University and studied under the authoritative scholar in the field of quantum computing, Yoshihisa Yamamoto. COO Ma Yin has worked in the aerospace industry for many years and was responsible for the design of precision instrument systems of several manned spacecraft and the Chinese space station.
There is a reason for the continuous injection of capital. Bose Quantum's quantum computing ability has found real application scenarios in the industry. The company said that it has "fully integrated quantum computing into the scientific research paradigm" and achieved more than 100 scenario explorations and applications in more than 20 industries such as life science, artificial intelligence, and communication.
Especially in the field of life science, quantum computing can play a role because life processes themselves follow the rules of quantum mechanics. Protein folding, enzyme - catalyzed reactions, the binding of drug molecules to targets, etc., all involve the interaction of a large number of electrons. Classical computers have extremely high computational complexity when simulating these processes and often need to compromise between accuracy and efficiency. However, quantum computers can directly simulate molecular orbitals and electronic structures in principle by using the characteristics of superposition and entanglement.
Ma Yin once introduced that in the field of biomedicine, Bose Quantum has achieved multiple "industry - university - research - application" cooperations. For example, the company cooperated with the Guangzhou National Laboratory to use the self - developed quantum Boltzmann machine for mRNA vaccine sequence design and optimization; it also cooperated with Shanghai Jiao Tong University to conduct research on molecular similarity calculation.
In mid - April, at the "Quantum Computing + AI for Science" application seminar hosted by Bose Quantum, the company shared more application cases in the fields of life science and medical health, such as bioinformatics, brain - computer interfaces, and organoids.
Take the application of quantum computing technology in the field of precision tumor treatment as an example: The cooperation between Bose Quantum and the team of Xiang Dongxi from the Shanghai Cancer Institute is a typical practice of deeply combining quantum computing power with clinical needs based on the coherent optical quantum computing platform.
From a technical principle perspective, many core problems in tumor diagnosis and treatment are essentially high - dimensional combinatorial optimization problems. Simply put, when we need to find the optimal solution from a large amount of complex data, the number of possible combinations will increase explosively as the number of variables increases. Traditional computers often take an extremely long time or even cannot complete the task when dealing with this scale.
Take the determination of intraoperative surgical margins as an example. The junction area between tumors and normal tissues is characterized by the aggregation of immunosuppressive signals and the enrichment of immune - escaping cells. Such microscopic differences cannot be accurately identified by traditional morphological detection. Therefore, Xiang Dongxi's team used Bose Quantum's 1000 - qubit coherent optical quantum computer to integrate the spatial location information of tumors and transcriptome matrix data and introduced the concept of "energy determination", that is, using the quantum computer to identify the different energy signal characteristics between normal tissues, tumor parenchyma, and the junction area between the two, and then construct a surgical margin determination model.
This process relies on the ability of the coherent optical quantum computer to solve the "Max - Cut" problem. Max - Cut is a classic problem in the field of combinatorial optimization, that is, given multiple nodes and connection relationships, how to divide them into two groups so that the sum of the weights of the cut connections is the largest. For classical computers, the higher the data dimension, the more difficult it is to solve this problem, showing an exponential growth. However, a dedicated 1000 - qubit quantum computer can complete the solution in milliseconds, with performance tens of thousands of times better than classical computing, making it possible to accurately identify residual tumors at the microscopic surgical margins during surgery.
In addition, in the cutting - edge field of brain science/brain - computer interfaces, quantum computing is also moving from theoretical concepts to practical verification platforms. The cooperation between Bose Quantum and the team of Sun Liuyang from the Shanghai Institute of Microsystem and Information Technology of the Chinese Academy of Sciences provides such an entry point.
One of the core challenges of brain - computer interfaces lies in the extremely high computational load of brain signal processing. The human brain has about 86 billion neurons. When traditional computers analyze these massive parallel data, decoding delay and data throughput are always insurmountable bottlenecks.
At this seminar, Sun Liuyang introduced the quantum optical computing neural decoding system jointly developed by the Shanghai Institute of Microsystem and Bose Quantum, which compressed the decoding delay of electroencephalogram signals to about 0.075 milliseconds. The processing delay of the traditional GPU solution is usually in the millisecond range, and 0.075 milliseconds means it is an order of magnitude faster. More importantly, this system has "scalability with constant complexity", that is, as the number of neural signal channels to be monitored increases from hundreds to thousands or even tens of thousands, the processing delay will not increase exponentially.
It is reported that from a technical path perspective, the key to this breakthrough lies in the integration of optical quantum computing and optical neural regulation technology. Another research direction of Sun Liuyang's team is non - invasive optical neural regulation, which uses the "near - infrared light penetration + photoelectric conversion" mechanism. It converts light signals into local electrical stimulation through up - conversion nanoparticles and photovoltaic materials, enabling deep - brain region regulation without electrode implantation. When the transmission of light pulse signals is combined with the optical quantum characteristics of quantum computing, the quantum computer can directly complete signal analysis in the "optical domain" without repeated photoelectric signal conversion, thus significantly reducing the delay. This is also the underlying logic for the quantum optical computing neural decoding system to achieve a 0.075 - millisecond delay.