Shaping the future of medical data analysis

The Smart Health Center at VinBigdata aims to build the VinDr ecosystem that revolutionizes the storage, processing, analysis, and understanding of medical data.

Request Demo

VinDr CAD

VinDr CAD is a platform for medical image analysis that consists of multiple Computer-Aided Diagnosis (CAD) tools to assist doctors in making fast and precise diagnoses. The platform can be seamlessly integrated into any Picture Archiving and Communication System (PACS) without breaking standard clinical workflows. Each CAD tool can automatically suggest diagnosis and localize a certain number of abnormalities based on appropriate DICOM images in a real-time fashion. Focusing on some of the most common imaging modalities, VinDr CAD currently offers 7 following tools.

VinDr-ChestXR

VinDr-ChestXR is a CAD tool for chest X-ray interpretation. It is able to identify 6 lung diseases and localize 22 types of common abnormalities on chest X-ray.

The system has been trained and validated on half a million chest X-ray studies from both public sources and several hospitals in Vietnam. The bounding box annotation and disease labeling for our private dataset have been performed by top Vietnamese radiologists. The accuracy of the system is above 90% for almost all diseases and findings.

VinDr-SpineXR

VinDr-SpineXR is a CAD tool for spine x-ray interpretation. It is able to classify a spine X-ray scan as normal and abnormal. The system can also localize 6 types of common abnormalities on the image.

VinDr-SpineXR has been trained and validated on a large-scale dataset of approximately 10 000 studies collected from several hospitals in Vietnam, in which the bounding box annotation and disease labeling have been performed by top Vietnamese radiologists. The accuracy for the abnormality detection is about 60% in terms of mAP@0.2.

VinDr-Mammo

VinDr-Mammo is a CAD tool for mammography interpretation. It is able to classify a mammography study into 3 BI-RADS (Breast Imaging-Reporting and Data System) levels and 4 types of breast density. The system can also localize 13 types of common abnormalities on mammography.

VinDr-Mammo has been trained and validated on about 50,000 studies collected from several hospitals in Vietnam. The bounding box annotation and disease labeling for this private dataset have been performed by top Vietnamese radiologists. The accuracy for BI-RADS classification is above 80%.

VinDr-ChestCT

VinDr-ChestCT is a CAD tool for Chest CT interpretation. It is able to identify 6 thoracic diseases and localize 24 types of common abnormalities on Chest CT scans.

The system will be trained and validated on about 30,000 studies collected from both public sources and several hospitals in Vietnam. The bounding box annotation (on 3-D volumes) and disease labeling for our private dataset have been performed by top Vietnamese radiologists.

VinDr-LiverCT

VinDr-LiverCT is a CAD tool for abdomen CT interpretation. It is able to identify 10 liver diseases, including different types of liver cancer, and localize 24 types of common liver abnormalities on abdomen CT scans.

The system will be trained and validated on about 10,000 studies collected from both public sources and several hospitals in Vietnam. The bounding box annotation (on 3-D volumes) and disease labeling for our private dataset have been performed by top Vietnamese radiologists.

VinDr-BrainCT

VinDr-BrainCT is a CAD tool for brain CT interpretation. It is able to identify 9 brain diseases, including several types of stroke, and localize 17 types of common abnormalities on brain CT scans.

VinDr-BrainMR

VinDr-BrainMR is a CAD tool for brain MRI interpretation. It is able to identify 9 brain diseases, including brain tumor, and localize 20 types of common abnormalities on brain CT scans.

The system will be trained and validated on about 3,000 studies collected from both public sources and several hospitals in Vietnam. The bounding box annotation (on 3-D volumes) and disease labeling for our private dataset have been performed by top Vietnamese radiologists.

Research

We are passionate about applying computer vision (CV), machine learning (ML) and deep learning (DL) models to build computed-aided detection (CAD) and computer aided diagnosis (CADx) systems from very large-scale clinical datasets of multiple imaging modalities (X-ray, CT, MRI, etc). Our research includes new methods and ML/DL models for radiologist-level understanding and interpretation from medical images. We aim to validate and publish our work on top-tier journals and conferences.

Medical Imaging

VinDr-SpineXR: A deep learning framework for spinal lesions detection and classification from radiographsEnhancing MRI Brain Tumor Segmentation with an Additional Classification Network

Hieu T. Nguyen, Hieu H. Pham, Nghia T. Nguyen, Ha Q. Nguyen, Thang Q. Huynh, Minh Dao, Van Vu – International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021).

Learning to Automatically Diagnose Multiple Diseases in Pediatric Chest Radiographs Using Deep Convolutional Neural Networks

Thanh T. Tran, Hieu H. Pham, Thang V. Nguyen, Tung T. Le, Hieu T. Nguyen, Ha Q. Nguyen – IEEE/CVF International Conference on Computer Vision Workshops (ICCV Workshop 2021)

DICOM Imaging Router: An Open Deep Learning Framework for Classification of Body Parts from DICOM X-ray Scans

Hieu H. Pham, Dung V. Do, Ha Q. Nguyen – IEEE/CVF International Conference on Computer Vision Workshops (ICCV Workshop 2021)

VinDr-RibCXR: A Benchmark Dataset for Automatic Segmentation and Labeling of Individual Ribs on Chest X-rays

Hoang C. Nguyen, Tung T. Le, Hieu H. Pham, Ha Q. Nguyen – International Conference on Medical Imaging with Deep Learning (MIDL 2021).

Enhancing MRI Brain Tumor Segmentation with an Additional Classification Network

Hieu T. Nguyen, Tung T. Le, Thang V. Nguyen and Nhan T. Nguyen – The 6th International Workshop on Brain Lesions 2020, MICCAI 2020, Peru, October (2020).

VinDr-CXR: An open dataset of chest X-rays with radiologist’s annotations

Ha Q. Nguyen, Khanh Lam, Linh T. Le, Hieu H. Pham, Dat Q. Tran, Dung B. Nguyen, Dung D. Le, Chi M. Pham, Hang T. T. Tong, Diep H. Dinh, Cuong D. Do, Luu T. Doan, Cuong N. Nguyen, Binh T. Nguyen, Que V. Nguyen, Au D. Hoang, Hien N. Phan, Anh T. Nguyen, Phuong H. Ho, Dat T. Ngo, Nghia T. Nguyen, Nhan T. Nguyen, Minh Dao, Van Vu – arXiv preprint.

Deep learning for detection and segmentation of artefact and disease instances in gastrointestinal endoscopy

Sharib Ali, Mariia Dmitrieva, Noha Ghatwary, Sophia Bano, Gorkem Polat, Alptekin Temizel, Adrian Krenzer, Amar Hekalo, Yun Bo Guo, Bogdan Matuszewski, Mourad Gridach, Irina Voiculescu, Vishnusai Yoganand, Arnav Chavan, Aryan Raj, Nhan T. Nguyen, Dat Q. Tran, Le Duy Huynh, Nicolas Boutry, Shahadate Rezvy, Haijian Chen, Yoon Ho Choi, Anand Subramanian, Velmurugan Balasubramanian, Xiaohong W. Gao, Hongyu Hu, Yusheng Liao, Danail Stoyanov, Christian Daul, Stefano Realdon, Renato Cannizzaro, Dominique Lamarque, Terry Tran-Nguyen, Adam Bailey, Barbara Braden, James East, Jens Rittscher – Medical Image Analysis Volume 70, May 2021.