An accurate and explainable deep learning system improves interobserver agreement in the interpretation of chest radiograph

Interpretation of chest radiographs (CXR) is a difficult but essential task for detecting thoracic abnormalities. Recent artificial intelligence (AI) algorithms have achieved radiologist-level performance on various medical classification tasks. However, only a few studies addressed the localization of abnormal findings from CXR scans, which is essential in explaining the image-level classification to radiologists. Additionally, the actual impact of AI algorithms on the diagnostic performance of radiologists in clinical practice remains…
Free Download
0 downloads

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

Most of the existing chest X-ray datasets include labels from a list of findings without specifying their locations on the radiographs. This limits the development of machine learning algorithms for the detection and localization of chest abnormalities. In this work, we describe a dataset of more than 100,000 chest X-ray scans that were retrospectively collected from two major hospitals in Vietnam. Out of this raw data, we release 18,000 images…
Free Download
10 downloads

A clinical validation of VinDr-CXR, an AI system for detecting abnormal chest radiographs

Computer-Aided Diagnosis (CAD) systems for chest radiographs using artificial intelligence (AI) have recently shown a great potential as a second opinion for radiologists. The performances of such systems, however, were mostly evaluated on a fixed dataset in a retrospective manner and, thus, far from the real performances in clinical practice. In this work, we demonstrate a mechanism for validating an AI-basedsystem for detecting abnormalities on X-ray scans, VinDr-CXR, at the…
Free Download
0 downloads

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

A wide range of diagnostic tasks can benefit from an automatic system that is able to segment and label individual ribs on chest X-ray (CXR) images. To this end, traditional approaches (Candemir et al., 2016) exploited hand-crafted features to identify the ribs, but failed with anterior ribs. Recently, deep learning (DL) has shown superior performance to other methods in the segmentation and labeling of individual ribs (Wessel et al., 2019).…
Free Download
3 downloads
0 downloads

An accurate and explainable deep learning system improves interobserver agreement in the interpretation of chest radiograph

Interpretation of chest radiographs (CXR) is a difficult but essential task for detecting thoracic abnormalities. Recent artificial intelligence (AI) algorithms have achieved radiologist-level performance on various medical classification tasks. However, only a few studies addressed the localization of abnormal findings from CXR scans, which is essential in explaining the image-level classification to radiologists. Additionally, the actual impact of AI algorithms on the diagnostic performance of radiologists in clinical practice remains relatively unclear. To bridge these gaps, we developed an explainable deep learning system called VinDr-CXR that can classify a CXR scan into multiple thoracic diseases and, at the same time, localize most types of critical findings on the image. VinDr-CXR was trained on 51,485 CXR scans with radiologist-provided bounding box annotations. It demonstrated a comparable performance to experienced radiologists in classifying 6 common thoracic diseases on a retrospective validation set of 3,000 CXR scans, with a mean area under the receiver operating characteristic curve (AUROC) of 0.967 (95% confidence interval [CI]: 0.958–0.975). The sensitivity, specificity, F1-score, false-positive rate (FPR), and false-negative rate (FNR) of the system at the optimal cutoff value were 0.933 (0.898–0.964), 0.900 (0.887–0.911), 0.631 (0.589–0.672), 0.101 (0.089– 0.114) and 0.067 (0.057–0.102), respectively. For the localization task with 14 types of lesions, our free-response receiver operating characteristic (FROC) analysis showed that the VinDr-CXR achieved a sensitivity of 80.2% at the rate of 1.0 false-positive lesion identified per scan. A prospective study was also conducted to measure the clinical impact of the VinDr-CXR in assisting six experienced radiologists. The results indicated that the proposed system, when used as a diagnosis supporting tool, significantly improved the agreement between radiologists themselves with an increase of 1.5% in mean Fleiss’ Kappa. We also observed that, after the radiologists consulted VinDr-CXR’s suggestions, the agreement between each of them and the system was remarkably increased by 3.3% in mean Co-hen’s Kappa. Altogether, our results highlight the potentials of the proposed deep learning system as an effective assistant to radiologists in clinical practice. Part of the dataset used for developing the VinDr-CXR system has been made publicly available at https://physionet.org/content/vindr-cxr/1.0.0/.

Download
10 downloads

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

Most of the existing chest X-ray datasets include labels from a list of findings without specifying their locations on the radiographs. This limits the development of machine learning algorithms for the detection and localization of chest abnormalities. In this work, we describe a dataset of more than 100,000 chest X-ray scans that were retrospectively collected from two major hospitals in Vietnam. Out of this raw data, we release 18,000 images that were manually annotated by a total of 17 experienced radiologists with 22 local labels of rectangles surrounding abnormalities and 6 global labels of suspected diseases. The released dataset is divided into a training set of 15,000 and a test set of 3,000. Each scan in the training set was independently labeled by 3 radiologists, while each scan in the test set was labeled by the consensus of 5 radiologists. We designed and built a labeling platform for DICOM images to facilitate these annotation procedures. All images are made publicly available in DICOM format in company with the labels of the training set. The labels of the test set are hidden at the time of writing this paper as they will be used for benchmarking machine learning algorithms on an open platform.

Download
0 downloads

A clinical validation of VinDr-CXR, an AI system for detecting abnormal chest radiographs

Computer-Aided Diagnosis (CAD) systems for chest radiographs using artificial intelligence (AI) have recently shown a great potential as a second opinion for radiologists. The performances of such systems, however, were mostly evaluated on a fixed dataset in a retrospective manner and, thus, far from the real performances in clinical practice. In this work, we demonstrate a mechanism for validating an AI-based
system for detecting abnormalities on X-ray scans, VinDr-CXR, at the Phu Tho General Hospital–a provincial hospital in the North of Vietnam. The AI system was directly integrated into the Picture Archiving and Communication System (PACS) of the hospital after being trained on a fixed annotated dataset from other sources. The performance of the system was prospectively measured by matching and comparing the AI results with the radiology reports of 6,285 chest X-ray examinations extracted from
the Hospital Information System (HIS) over the last two months of 2020. The normal/abnormal status of a radiology report was determined by a set of rules and served as the ground truth. Our system achieves an F1 score–the harmonic average of the recall and the precision–of 0.653 (95% CI 0.635, 0.671) for detecting any abnormalities on chest X-rays. Despite a significant drop from the in-lab performance, this result establishes a high level of confidence in applying such a system in real-life situations.

Download
3 downloads

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

A wide range of diagnostic tasks can benefit from an automatic system that is able to segment and label individual ribs on chest X-ray (CXR) images. To this end, traditional approaches (Candemir et al., 2016) exploited hand-crafted features to identify the ribs, but failed with anterior ribs. Recently, deep learning (DL) has shown superior performance to other methods in the segmentation and labeling of individual ribs (Wessel et al., 2019). However, developing DL algorithms for this task requires annotated images for each rib structure at pixel-level. To the best of our knowledge, there exists no such benchmark
datasets and protocols. Hence, we present VinDr-RibCXR – a benchmark dataset for the automatic segmentation and labeling of individual ribs on CXRs. This work also reports performance of several state of-the-art DL-based segmentation models on the VinDr-RibCXR dataset. The dataset and codes will be made publicly available to encourage new advances.

Download
1 download

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

X-ray imaging in Digital Imaging and Communications in Medicine (DICOM) format is the most commonly used imaging modality in clinical practice, resulting in vast, nonnormalized databases. This leads to an obstacle in deploying artificial intelligence (AI) solutions for analyzing medical images, which often requires identifying the right body part before feeding the image into a specified AI model. This challenge raises the need for an automated and efficient approach to classifying body parts from X-ray scans.

Download
3 downloads

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

Chest radiograph (CXR) interpretation is critical for the diagnosis of various thoracic diseases in pediatric patients. This task, however, is error-prone and requires a high level of understanding of radiologic expertise. Recently, deep convolutional neural networks (D-CNNs) have shown remarkable performance in interpreting CXR in adults. However, there is a lack of evidence indicating that D-CNNs can recognize accurately multiple lung pathologies from pediatric CXR scans.

Download

Case Studies

The comprehensive Vietnamese virtual assistant on VinFast’s electric car

Voice assistant ViVi is an intelligent voice control system developed by VinBigData (Vingroup), applying cutting-edge technologies such as artificial intelligence (AI), big data analysis, natural language processing, etc. ViVi has an outstanding advantage of native language, with the ability to recognize Vietnamese with 98% accuracy (for common words) and understand the multi-regional Vietnamese language in…

vinocr tích hợp trên cihms
logo vinpearl

To enhance the experience of consulting services in hospitality with Vinpearl

Service provided: Support consulting and sales process on VinPearl’s websites and other channels for customer interaction Business Demand: Increase customer experience on online channels Prioritize sales process automation Reduce work for customer service staff. Integration: Website/Facebook/Zalo with more than 6 languages Integrate with other systems of VinPearl CRM/Service Booking/Survey/Promotion/Hotel Management/etc.

Combining the power of technology, bringing consumers into the new digital era

Service provided: creating chatbot to support customer service of VinSmart Business Demand: Support in solving business situations relating to smartphones or TV such as: Warranty points Price of replacement components User manual guide Receive and respond to complaints relating to hardware and software errors Integration: Facebook & Website of VinSmart

vinbot tại vinmec

Vinmec – More passionate customer service with VinBot

Service provided: creating chatbot to support services of Vinmec Business Demand: Support in solving business situations such as: Hospital information Product price Consulting service packages Booking appointments for health check Receive and respond to patient complaints Integration: Facebook, Website and App of Vinmec (OneVinmec)

AI Camera Grand World Phú Quốc
logo vinpearl

Advanced AI Camera System creates a Smart Resort Complex – Grand World Phu Quoc

Advanced AI Camera System optimizes business operations and automates customer management cycle at the leading smart resort complex in Southeast Asia – Grand World Phu Quoc. Scale AI Camera Monitoring System with intelligent features and unlimited scalability Total area of Grand World Phu Quoc: 800.000m2​ Average No. of Visitors 20.000 visitors/day Detail Description Hardware installation…

VinDr AI is an effective “assistant” for radiologists at Hanoi Medical University Hospital

Developed based on a dataset of nearly half a million studies with detailed lesion labels and disease conclusions from major hospitals in Vietnam, VinDr AI helps radiologists accurately detect and localize many types of abnormalities in chest, mammo and spine. Meanwhile, to provide doctors the best support, the solution also marks the studies by color:…

Manage customers easily and quickly with AI Camera System at Vincom Times City Shopping Mall

Scale Total area of Vincom Times City: 200.000m2​ Average No. of Visitors 126.000 visitors/day Project Detail Supplied Service & Product​ Visitors Counting & Analyzing system at Vincom: Total visitors at different areas​ Identify & measure the number of old/new visitors​ Determine peak time frame of specific areas.​ Determine the number of customers at each area…

camera thông minh
camera thông minh

Smart surveillance cameras in 3 Vinhomes megacities

Smart Camera System is deployed in 3 leading megacities: Vinhomes Grand Park, Vinhomes Ocean Park and Vinhomes Smart City to support security and safety for residents living in buildings. Applying advanced technologies to identify and track the movement of vehicles and objects and analyze object recognition characteristics, the system can operate in all different environmental…

tính năng OCR

VinOCR helps check-in without receptionists, promoting CiHMS hotel management solution

The OCR feature, developed by VinBigData, is integrated with CiHMS hotel management solution to “digitize information”, unifying data from all hotel management and operation. Project Detail Implement VinOCR integrated with CiHMS hotel management application on smartphones, helping to digitally transform the processes of handling paperwork and managing hotel customer information on the system with a…

    Thank you for your interest.

    File hiện tại không thể tải xuống
    Vui lòng liên hệ hỗ trợ.

    VinOCR eKYC

    Chọn ảnh từ máy của bạn

    Chọn ảnh demo dưới đây hoặc tải ảnh lên từ máy của bạn

    Tải lên ảnh CMND/CCCD/Hộ chiếu,...

    your image
    Chọn ảnh khác
    Tiến hành xử lý

    Thông tin đã được xử lý

    Mức độ tin cậy: 0%

    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -
    • -

    Xác thực thông tin thẻ CMND/CCCD

    Vui lòng sử dụng giấy tờ thật. Hãy đảm bảo ảnh chụp không bị mờ hoặc bóng, thông tin hiển thị rõ ràng, dễ đọc.

    your image
    Chọn ảnh khác

    Ảnh mặt trước CMND/CCCD

    your image
    Chọn ảnh khác

    Ảnh mặt sau CMND/CCCD

    your image
    Chọn ảnh khác

    Ảnh chân dung

    This site is registered on wpml.org as a development site.