About Us

Keya Medical develops artificial intelligence applications that help healthcare delivery organizations provide better patient care. Our R&D team is made up of over 70 people, including doctors and post-doctors from leading universities such as Yale, Cornell, Duke, and the University of North Carolina.

Keya Medical conducts scientific research collaborations with more than 200 hospitals around the world. The main applications of our products include cardiovascular function assessment, medical imaging assisted diagnosis, and clinical decision support. Keya Medical’s flagship product, DEEPVESSEL FFR, is the only CT-derived FFR product approved for clinical use in China. It was the first deep learning medical image analysis product in China to obtain CE certification.

AI for Smarter Healthcare

Combining advanced deep learning algorithms and other core technologies with advanced medical concepts, we provide accurate diagnosis and treatment solutions to hospitals, physical examination centers, third-party imaging centers and other medical institutions. Keya Medical collaborates with healthcare institutions to develop and to demonstrate the clinical effectiveness of advanced medical AI solutions. Today, Keya Medical products are widely used in Chinese hospitals, coronary catheterization labs, and imaging centers.

Core Competence

Since 2016, Keya Medical’s strategy has been to assemble a world-class leadership team. Using a powerful cloud-based AI platform capable of processing large volumes of data with advanced AI algorithms, the company continuously updates its knowledge base. Our DEEPVESSEL FFR product has passed a number of clinical trials to validate its accuracy. Keya Medical has also developed expertise in regulatory science, having achieved successful clearances and approvals including a CE Mark, FDA 510(k) clearance, and NMPA Class-III certification for different products.

Our Achievements

Prior to founding Keya Medical, our team of data scientists, engineers, and designers received educations from leading academic institutions around the world. Many of our scientists developed advanced AI technologies for top institutions around the world.

We are committed to developing solutions that solve real clinical challenges. To ensure that AI software meets real needs, we work in collaboration with clinicians at over 200 health systems around the world to understand workflow, end-user needs, and other human factors. Our strong focus on the collaborative development of new AI technologies differentiates us from other AI companies.


Keya Medical is founded


Keya Medical establishes the first cardiovascular AI diagnostic laboratory in China


DEEPVESSEL FFR enters the “Special Approval Channel for Innovative Medical Devices” by the CFDA

Keya Medical receives ISO 13485 certification

DEEPVESSEL FFR receives CE certification


DEEPVESSEL FFR becomes the first Class-III medical device approved by the NMPA


Keya Medical publishes COVID-19 research results in the journal, Radiology


Artificial Intelligence Distinguishes COVID-19 from Community Acquired Pneumonia on Chest CT
Li L., Qin L., Xu Z. Yin, Y., Wang, X., et al.
Radiology, March 2020.

Learning Physical Properties in Complex Visual Scenes: An Intelligent Machine for Perceiving Blood Flow Dynamics from Static CT Angiography Imaging
Gao Z., Wang, X., Sun, S., Wu, D., Bai, J., Yin, Y., et al.
Neural Networks, 2020.

Learning Tree-structured Representation for 3D Coronary Artery Segmentation
Kong, B., Wang, X., Bai, J., Lu, Y., Gao, F., Cao, K., et al.
Computerized Medical Imaging and Graphics, 2020.

DeepCenterline: A Multi-task Fully Convolutional Network for Centerline Extraction
Guo Z., Bai, J., Lu, Y., Wang X., Cao K., et al.,
International Conference on Information Processing in Medical Imaging (IPMI), 2019.

Evaluation of Fractional Flow Reserve in Patients with Stable Angina: Can CT Compete with Angiography?
Liu, X., Wang, Y., Zhang, H., Yin, Y., Cao, K., et al.
European Radiology, 2019.

Automated Anatomical Labeling of Coronary Arteries via Bidirectional Tree LSTMs
Wu, D., Wang, X., Bai, J., Xu, X., Ouyang B., et al.
International Journal of Computer Assisted Radiology and Surgery, 2019.14: 271.

Precise Diagnosis of Intracranial Hemorrhage and Subtypes Using a Three-dimensional Joint Convolutional and Recurrent Neural Network
Ye H.*, Gao F.* (*Equal contribution), Yin, Y., Guo D., Zhao, P., et al.
European Radiology, March 2019.

Diagnostic Accuracy of a Deep Learning Approach to Calculate FFR from Coronary CT Angiography
LWang, Z., Zhou, Y., Zhao, Y., Shi, D., Liu, Y., et al.
Journal of Geriatric Cardiology, 2019.

Train a 3D U-Net to Segment Cranial Vasculature in CTA Volume without Manual Annotation
Chen, X., Lu, Y., Bai, J., et al.
IEEE 15th International Symposium on Biomedical Imaging (ISBI), 2018.

Invasive Cancer Detection Utilizing Compressed Convolutional Neural Network and Transfer Learning
Kong, B., Sun, S., Wang, X., et al.
In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2018.

Integrate Domain Knowledge in Training CNN for Ultrasonography Breast Cancer Diagnosis
Liu, J., Li, W., Zhao, N., Cao, K., Yin Y., et al.
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2018.

Hemodynamics Analysis of the Serial Stenotic Coronary Arteries
Liu, X., Peng, C., Xia, Y., et al.
Biomedical Engineering Online.16(1).127, 2017.

A Study of Noninvasive Fractional Flow Reserve Derived from a Simplified Method based on Coronary Computed Tomography Angiography in Suspected Coronary Artery Disease
Shi, C., Zhang, D., Cao, K., et al.
Biomedical Engineering Online.16(1).43, 2017.

Risk Stratification of Lung Nodules Using 3D CNN-Based Multi-task Learning
Hussein, S., Cao, K., Song, Q., et al.
International Conference on Information Processing in Medical Imaging (IPMI), 2017.

Cancer Metastasis Detection via Spatially Structured Deep Network
Kong B*, Wang X* (*Equal contribution), Li Z, Song Q, Zhang S.
International Conference on Information Processing in Medical Imaging (IPMI), 2017.

TumorNet: Lung Nodule Characterization using Multi-view Convolutional Neural Network with Gaussian Process
Hussein, R. Gillies, K. Cao, et al.
IIEEE 14th International Symposium on Biomedical Imaging (ISBI), 2017.