With the rising incidence of hypertension and the accelerating trend of population aging, the annual number of new stroke cases continues to increase, among which ischemic stroke accounts for the highest proportion. Clinically, the most effective treatments for ischemic stroke are thrombolysis and thrombectomy.
According to the Chinese Clinical Practice Guidelines for Stroke, without imaging evaluation, thrombolysis and thrombectomy can only be performed on patients whose onset time is within the "golden treatment window" (4.5 hours for thrombolysis and 6 hours for thrombectomy). However, in 2019, the international multicenter clinical studies DIFFUSE-3 and DAWN showed that among patients whose onset time exceeded the "golden treatment window", those who met specific imaging criteria could also receive thrombolysis and thrombectomy. These imaging criteria need to be measured by automated analysis software.
Bian Yueyan is the founder and CEO of Nanjing Yuexi Medical Technology Co., Ltd. (hereinafter referred to as "Yuexi Medical"). The cover of his WeChat Moments features Rukawa Kaede, the protagonist of the classic anime Slam Dunk—a character who always faces strong opponents with passionate fighting spirit and seeks opportunities to break through in every attack.
In January 2020, Bian Yueyan founded Yuexi Medical and resolutely entered the emerging field of AI medical imaging, focusing on stroke—a disease with the highest mortality and disability rates in China. In just two years, the company successfully developed its first "full lifecycle" health management service product for ischemic stroke. In clinical stroke diagnosis services, this product achieves an imaging analysis accuracy rate of 98.6% and an average processing time of only 1.5 minutes.
Looking back on the five-year entrepreneurial journey, it is like a long basketball game: every difficulty is a tight double-team encountered, and every decision is crucial to winning or losing. Bian Yueyan firmly believes that only by maintaining a steady pace with firm faith and seeking gaps with an optimistic attitude can we complete that decisive breakthrough layup.
Unlocking the Burden of Image Reading with the Key of AI
When Bian Yueyan first had the idea of starting a business, he had already worked in the field of medical imaging for many years.
During his master’s studies in biomedical engineering at Imperial College London, his main research direction was neuroimaging, which laid a solid academic foundation and a cutting-edge technical perspective for him.
After returning to China, he joined United Imaging Healthcare and Neusoft Medical—well-known domestic digital diagnosis and treatment equipment service providers—serving as a clinical product department manager and senior algorithm engineer. Notably, in his role as the project leader of the National Thrombolysis and Thrombectomy Imaging Platform jointly built by Neusoft Medical and Xuanwu Hospital, he participated in the formulation and acceptance of the National Thrombolysis and Thrombectomy Imaging Standards under the special work of the Beijing Cerebrovascular Disease Industry Technology Innovation Strategic Alliance, and obtained more than 10 national invention patents.
During this period, Bian Yueyan also truly felt the heavy pain points in clinical practice. Through visits to hospital radiology departments, he found that a radiologist needs to complete CT diagnoses for at least 60 patients every day. Each patient’s imaging data includes up to 250-300 images, and doctors have to review each patient’s imaging data repeatedly 3-4 times.
This means that for each patient, a doctor needs to view images thousands of times, and the total number of image views in a workday may reach tens of thousands or even hundreds of thousands. The entire diagnosis process relies heavily on personal experience and concentration, leading to a high risk of misdiagnosis and missed diagnosis.
At the same time, relevant medical data shows that more than 90% of medical data comes from medical imaging. On one side, the value of massive data is yet to be tapped; on the other side, doctors are burdened with excessive image-reading pressure and the resulting diagnostic risks. Standing at the crossroads where biomedical engineering and information technology converge, Bian Yueyan clearly saw that "vital opportunity"—AI is the key to breaking free from the industry’s predicament and moving toward a new future of medical imaging.
"The core value of AI technology lies in its ability to efficiently integrate and analyze massive imaging data through deep learning. This not only enables accurate prediction of disease risks and intelligent auxiliary localization of lesions, fundamentally reducing the risk of missed diagnosis and misdiagnosis, but also frees doctors from heavy repetitive work, comprehensively improving the efficiency and quality of diagnosis and treatment," said Bian Yueyan.
Only through interactions with like-minded people can ideas spark brighter sparks. During that period, while devoting himself to the research and development of key technologies, Bian Yueyan also continuously shared insights and discussed the future with friends around him, constantly calibrating the direction of his entrepreneurship and anticipating potential challenges ahead. It was in this process that he attracted many partners who also had a passion for entrepreneurship and a desire to create real social value.
In 2020, a core team of five people was assembled. All members graduated from well-known universities at home and abroad, including the University of Southern California (USA), Tsinghua University (China), Eindhoven University of Technology (the Netherlands), and Blekinge Institute of Technology (Sweden). As a symbol of the team’s spirit, they integrated the five members’ names into the company’s name, giving birth to "Yuexi".
After the company’s establishment, its strategic direction was quickly clarified: instead of providing large-scale, all-encompassing multi-disease services, it would focus on stroke. This disease ranks first in mortality and disability rates among Chinese residents, with nearly 300 million potential patients. However, more than half of the patients can recover through rehabilitation treatment. The growing number of stroke cases has driven a surge in demand for the AI stroke diagnosis medical device industry. Relevant research shows that from 2022 to 2024, the market size of China’s AI stroke diagnosis medical device industry grew from 180 million yuan to 610 million yuan, and is expected to reach 6.36 billion yuan by 2030.
Penetrating "Black Box" Doubts with Clinical Evidence
With a clear direction, the team unleashed remarkable energy, achieving "rapid" progress in research and development.
In 2022, the company launched imSTROKE, its first "full lifecycle" health management service product for ischemic stroke. It is also the only international automated imaging analysis product that covers the recommendations above Level IIB in the 2019 AHA/ASA Guidelines for the Early Management of Acute Ischemic Stroke. In June of the same year, imSTROKE obtained certification from China’s National Medical Products Administration (NMPA) and was approved for market launch.
Based on imSTROKE, Yuexi Medical has built an integrated "full lifecycle" health management service solution that combines medical artificial intelligence, clinical big data analysis, and cloud platform technology. This solution enables "full lifecycle" disease management services, covering risk early warning before disease onset, clinical diagnosis and decision-making during the disease, post-operative medication efficacy evaluation, and prevention of disease recurrence.
Taking stroke clinical diagnosis services as an example, verified by multicenter clinical trials, the product achieves an imaging analysis accuracy rate of 98.6%; the average processing and analysis time is only 1.5 minutes, with a maximum of no more than 3 minutes; it realizes full-process automated processing and analysis, and automatically generates imaging analysis reports with one click.
A real clinical case vividly demonstrates its value: a 66-year-old male patient was admitted to the hospital with slow response and limb movement disorders for 6.5 hours. Processed by imSTROKE, imaging analysis was completed in only 3 minutes, accurately calculating the volume of the core infarct area and ischemic penumbra, with a mismatch ratio that met the guideline criteria for thrombectomy. Based on this, the attending doctor made a decisive decision to perform thrombectomy on the patient 8 hours after onset, recanalizing the occluded left middle cerebral artery. After the operation, the patient’s limb movement function partially recovered, and fully returned to normal 24 hours later.
However, for a startup company, the advanced nature of technology and the approval of products are only the first step of a long march. Whether the product can achieve large-scale implementation is another practical hurdle lying ahead.
"The biggest difficulty comes from doubts about the product’s accuracy," Bian Yueyan admitted. "Artificial intelligence technology itself has a certain ‘black box’ attribute, and coupled with our team’s youth, doctors generally held a cautious and skeptical attitude when we initially contacted various hospitals."
To resolve hospitals’ doubts, Yuexi Medical took the initiative to propose on-site testing: using the imaging data of the cooperative hospitals themselves for case demonstrations. Since doctors are very familiar with the cases in their own hospitals, the product’s analytical capabilities and accuracy were intuitively verified. In addition, the company also collaborated with multiple hospitals to conduct clinical verification trials and published the research results in well-known domestic and international academic journals, providing strong academic endorsement for the product’s reliability.
In terms of market promotion, they adopted a "top-down" implementation strategy: first, co-establishing model pilot projects with leading hospitals in various regions, then using the demonstration effect and authority of these pilot hospitals to gradually expand to lower-level and primary medical markets.
This combination of "technical confidence + academic endorsement + flexible strategy" allowed Yuexi Medical to expand its market territory steadily and efficiently. Currently, the company has established scientific research and sales cooperation with more than 210 hospitals, covering multiple provinces, municipalities, and autonomous regions including Beijing, Shanghai, Guangdong, Jiangsu, Shandong, and Henan.
Navigating Entrepreneurial Risks and Expanding Application Scenarios
Looking back on the journey, Bian Yueyan sighed with emotion: "There are countless hardships and dangers in the entrepreneurial process. Only with firm faith and optimism can a turning point possibly appear before us."
In the early days of the company’s establishment, the huge R&D investment soon exhausted the team members’ own funds. Just when everyone was at a loss, they met an investor who highly recognized the company’s product concept and business model. The two parties hit it off immediately and finalized the seed round of investment within a few days, helping the company get through the most difficult period.
At the end of 2024, relying on its "full lifecycle" health management service for stroke, Yuexi Medical stood out in the 12th "Win in Nanjing" Youth College Student Entrepreneurship Competition and won the first prize. This award also brought it to the attention of Nanjing Venture Capital Group, and the company successfully obtained 500,000 yuan in investment and support funds from the group and Nanjing Municipal Human Resources and Social Security Bureau.
When talking about the investment logic, the relevant person in charge of the Future Angel Investment Department of Nanjing Venture Capital Group pointed out that with the increasing number of stroke cases and the growing demand for efficient and accurate diagnostic tools, the market size of AI diagnostic software in the stroke field will continue to expand. The Yuexi Medical team not only has a solid technical background but also possesses the ability to transform scientific research achievements into practical products. Currently, it has successfully delivered multiple practical application cases, and its products have achieved significant results in clinical verification at multiple stroke centers. "The progress of this project is impressive. This year, the project has successfully entered the procurement lists of many hospitals. For a startup company, being able to pass the strict procurement standards of hospitals is in itself a strong proof that its technical advancement and product innovation have been recognized," the person in charge said.
Today, Yuexi Medical has entered a path of healthy development and achieved stable profitability. Standing at a new starting point, Bian Yueyan is setting his sights on a more distant future.
On one hand, the company is promoting the extension of scenarios "from in-hospital to out-of-hospital". It has completed the R&D of a prototype of a home-use portable ultrasound screening device. Positioned as a daily brain health monitoring device, this product is as easy to use as a blood pressure monitor or blood glucose meter, allowing non-professionals to operate it easily. It can also connect to a cloud doctor platform to obtain professional interpretations, realizing seamless connection between screening and consultation.
On the other hand, the company is expanding disease coverage "from stroke to common brain diseases". With the support of 12,000 cases of multi-modal magnetic resonance imaging (MRI) from its cooperation with encephalopathy hospitals, the company is building a multi-classification deep learning model, extending AI imaging screening technology to common brain diseases such as Alzheimer’s disease, cerebral small vessel disease, neurogenic depression, and neurocognitive disorders, and striving to become a comprehensive guardian in the field of brain health.
Starting with technology, succeeding with faith, and ultimately arriving at the health expectations of thousands of households. On this new medical battlefield built by code and algorithms, a group of young people are using the most cutting-edge technology to guard the most fragile lives.
Source: Venture Capital Group
Review: Xue Yao
Release: You Yi
