Aitrox Technology and Sir Run Run Shaw Hospital Launch a New Era of Non-Invasive AI-Powered Cardiovascular Diagnosis

On May 14, 2019, the 81st China International Medical Equipment Fair (Spring) grandly opened at the National Exhibition and Convention Center (Shanghai). As Fosun’s first independently incubated artificial intelligence enterprise, Aitrox Technology showcased its full product lineup at this year’s CMEF Spring Exhibition.

Guided by its mission to “empower one billion households worldwide with healthy living through artificial intelligence,” Aitrox Technology has developed a series of internationally leading intelligent computer-aided diagnosis (CAD) products, addressing critical pain points in medical diagnostics. The two major product lines showcased at this event, “Aitrox Ruiying” and “Aitrox Jingling,” provide multimodal AI-assisted diagnostic solutions for multiple anatomical regions—including the chest, musculoskeletal system, and coronary arteries—covering more than 20 disease types. Upon their launch, these products demonstrated the highest global technological standards, not only directly resolving urgent challenges faced by radiology and technical departments but also offering valuable insights into future research directions for attending experts.

Among these, the latest product jointly developed by Aitrox Technology and the Department of Radiology at Sir Run Run Shaw Hospital in Hangzhou—the “One-Stop Intelligent Auxiliary Diagnosis Solution for Coronary Heart Disease”—was unveiled to the industry through public channels for the first time, attracting attention from healthcare administrators and radiology experts worldwide.

Second from the left:Director Hu Hongjie, Department of Radiology, Sir Run Run Shaw Hospital, Hangzhou

According to the "China Cardiovascular Disease Report 2018," there are currently 11 million patients with coronary heart disease (CHD) in China, with a mortality rate as high as 63%. The effective prevention and treatment of CHD has become a severe challenge facing both patients and healthcare professionals. Coronary angiography is the traditional "gold standard" for diagnosing CHD; however, this invasive procedure carries certain clinical risks and requires guidance from senior physicians. Additionally, the high cost of consumables such as guidewires makes the diagnostic examination expensive for patients, thereby hindering its widespread application in medically underdeveloped regions.

From a clinical perspective, it is difficult to determine whether coronary artery stenosis causes myocardial ischemia using DSA alone. Therefore, a diagnostic method using pressure guidewires to quantitatively assess the degree of myocardial ischemia has been proposed. However, due to high costs, invasiveness, and radiation exposure, this method has not been widely adopted in China.

However, the method of simulating FFR values via hemodynamic modeling, which is widely prevalent in North America, has significant drawbacks. For instance, each coronary case must be uploaded to the cloud for manual mesh generation and modeling by fluid dynamics engineers, using commercial simulation software. This process not only requires substantial manual intervention but also takes 4–6 hours to complete. Furthermore, uploading data to the cloud remains difficult to implement in hospitals within China.

Consequently, many domestic enterprises have adopted an alternative approach by using machine learning to fit hemodynamic results, thereby accelerating the entire workflow. While this is a promising strategy, the generalizability of such algorithms remains questionable due to the limited availability of training data in China, which amounts to only hundreds of cases per year.

To address the aforementioned clinical pain points, Aitrox Technology has partnered with the Department of Radiology at Sir Run Run Shaw Hospital in Hangzhou to launch a collaborative project on FFRct diagnostic products. By combining DSA data with CTA imaging for training, the project integrates deep learning techniques, hemodynamics, and other innovative approaches for in-depth research and development. Aligned with clinical needs and building upon the principles of computational fluid dynamics (CFD) simulation, Aitrox has independently developed a fully automated mesh generation system and a proprietary CFD solver. This innovation reduces the processing time from the original 4–6 hours to just 10 minutes, eliminates the need for involvement by fluid dynamics engineers, and ensures high accuracy.

With the FFRct AI diagnostic product, physicians can initiate intelligent segmentation of cardiac and coronary structures from patient CTA images with a single click, visually displaying conditions in various regions such as the atria and ventricles. As the mouse cursor moves, the AI system displays FFR (Fractional Flow Reserve) values at any location within the vessels, along with multiple parameters including blood pressure, blood flow, and wall shear stress. It also uses color coding to intuitively indicate the severity of myocardial ischemia, with the entire process taking approximately 10 minutes. Without requiring manual configuration by physicians, patients can receive diagnostic results for coronary stenosis within half an hour, ensuring seamless integration into clinical diagnosis and treatment workflows.

Dr. Hu Hongjie, Director of the Department of Radiology at Sir Run Run Shaw Hospital in Hangzhou, stated: “In addition to assessing coronary artery stenosis, clinical diagnosis using cardiac CTA imaging requires precise identification and quantitative evaluation of a series of indicators, including calcium scoring, myocardial function, and plaque composition. China possesses vast medical imaging data resources and extensive clinical experience among physicians, providing the capability and advantage to independently develop AI-based diagnostic products for FFRCT. By leveraging artificial intelligence technology, we can accelerate the comprehensive clinical adoption of this novel non-invasive screening method for coronary heart disease, truly benefiting a broad patient population and enabling early detection and treatment.”

In the future, Aitrox Technology will join forces with Sir Run Run Shaw Hospital and other medical institutions to deepen scientific research innovation. By integrating clinical applications with advanced technologies such as deep learning, clinical pathways, and physics, the company will conduct further explorations in disease diagnosis along the direction of precision diagnosis and treatment for coronary heart disease (CHD). The goal is to create a one-stop, clinically empowering diagnostic solution for CHD, providing patients with more timely, personalized, and high-quality treatment.

At this year's CMEF, Aitrox Technology showcased two product lines: "Aitrox Ruiying" and "Aitrox Jingling".

Xingmai Ruiying is an artificial intelligence system grounded in evidence-based medicine, leveraging deep learning, computer vision, and hemodynamics to assist in medical imaging diagnosis. Primarily designed for radiology departments, the product currently covers the analysis and auxiliary diagnosis of multi-system, multi-modal imaging data (including DR, CT, and MRI) across the head and neck, chest, cardiovascular and cerebrovascular systems, abdomen, and musculoskeletal system. It can also serve as a clinical tool for departments such as respiratory medicine, orthopedics, cardiology and neurology, and endocrinology, forming a comprehensive product portfolio.

Aitrox LensAI is an all-in-one AI-powered microscope independently developed by Aitrox Technology. By integrating optical imaging, automated control, and computer vision technologies, it enables high-speed panoramic sample scanning, high-quality digital image processing, and precise auxiliary diagnostic decision-making. It provides hospital pathology and clinical laboratory departments with a comprehensive, integrated smart diagnostic solution combining software and hardware, significantly enhancing medical digitalization, upgrading research infrastructure, supplementing diagnostic resources, and improving diagnostic accuracy and efficiency.

The latest product manuals have been published on the FUSUN AITROX official WeChat account (fosun-aitrox).

As the first independently incubated artificial intelligence enterprise under Fosun, Aitrox Technology is a high-tech company specializing in medical AI, boasting a “dream team” composed of AI elites and authoritative medical experts. Upholding its mission to “enable healthy living for one billion households worldwide through artificial intelligence,” the company drives a paradigm-shifting leap from evidence-based medicine to precision medicine through technological innovation.