Dataset Overview
Definition：Chest X-Ray Plain Film Imaging Datasetis provided byChangsha Langhui Information Technology Co., Ltd.A large chest X-ray image label database. Covers multiple chest tablets, including real (PA/AP), side, and maintains original DICOM images and exposure parameters, accompanied by a double-checked structured diagnostic report covering dozens of breast disease labels, such as pneumonia, tuberculosis, chest fluid, aerobic chest, pulmonary chords/prosy, heart enlargement, etc.
| Dataset Name | Chest X-Ray Plain Film Imaging Dataset |
| Total Data Volume | Total X-Ray Volume Included: 1,700,000 Cases |
| Imaging Modality | X-Ray (DR/CR) |
| Covered Body Parts | Chest / Chest Radiograph (PA/AP + Lateral) |
| Raw Data | De-identified original DICOM, retaining exposure parameters, pixel spacing, body position / view, device manufacturer and model |
| Diagnostic label | Pneumonia / pulmonary tuberculosis / pleural effusion / pneumothorax / pulmonary nodule / lung mass / enlarged cardiac silhouette / widened mediastinum, etc. |
| Review Mechanism | Double-Review Gold Standard (initial reader + reviewing physician) |
| Pre-Operative / Post-Operative | Post-Operative Data Accounts for ≤10% |
| Source Institution | Radiology Department of Partner Grade-A Tertiary Hospitals |
| Use Cases | Chest X-Ray AI-Assisted Diagnosis, Lung Disease Screening, Intelligent Emergency Triage, Multi-Label Classification |
Delivery Specifications and Field Requirements
The dataset strictly follows the XR-CHEST unified delivery index requirements, ensuring data quality and traceability。
Delivery & Counting：One row corresponds to one chest X-ray examination / Study；Multiple sequences / multiple views count as only 1 case
Coverage：Covering all visible structures on the chest radiograph, including lung fields, cardiac silhouette, mediastinum, diaphragm, costophrenic angle and bones
Positives and Composition：Definite lesions / positive examinations as the main body；Normal complete examinations, postoperative follow-up examinations and quality-limited examinations are grouped separately
Dedicated Fields：Body position, projection conditions, lesion location / size / morphology, distribution characteristics, complications and follow-up
Cross-Data Association：Prioritize linkage with chest CT, sputum tests / culture, pathology, treatment and follow-up
2026 Frontier AI Research Progress
Domain Review：In 2025-2026, chest X-ray AI focused on lightweight models and XAI explainability. LXNet achieved nine-class classification and CD-DETR optimized small lesion detection, while multi-label classification and generative data augmentation became mainstream directions。
LXNet Lightweight Chest X-Ray Lung Disease Classification
PLOS ONE, 2025The lightweight CNN achieves nine classification (normal/pneumonia/high/low density/obstructive pulmonary disease, etc.), 6,743 image data sets with the Grad-CAM interpretability.
CD-DETR Small Lesion Detection for Chest Diseases
PLOS ONE, 2025An optimized detection Transformer for small lesion detection in chest diseases, with multi-scale feature fusion improving detection accuracy。
EfficientNetB0 Multi-Label Chest X-Ray Classification
Scientific Reports, 2026In NIH data sets, multi-label classification using CLAHE comparison enhancement, category balance and migration learning is used to improve multi-disease detection capabilities.
RADAI Lung Abnormality Classification and Report Generation
2025A CNN-based model detects four types of lung abnormalities and automatically generates structured diagnostic reports, advancing the AI-assisted image reading workflow。
Annotation Workflow & Quality Control
Image Acquisition & De-identification
Standardized acquisition workflow: complete data is exported directly from the equipment, and patient identifying information (PHI) is removed before the data is stored。
Initial Annotation (Specialist Physician)
Attending physicians annotate case by case against the standards, including lesion localization, morphological description and disease term determination。
Review (Associate Chief Physician or Above)
Experts with the title of Associate Chief Physician or above review the initial annotation results item by item, correcting erroneous annotations and supplementing missing dimensions。
Consistency Assessment
10% of samples are randomly selected and independently annotated by 3 physicians to calculate Fleiss' Kappa; below 0.Dimensions scored 75 are flagged for rework。
License and Usage Agreement
Academic Research License
For universities and research institutions, supporting academic research related to medical AI。After signing the agreement, a de-identified data subset is provided; the source must be acknowledged：Langhui Technology DataAssetsAPI。
Commercial License
For medical device companies and AI diagnostics companies, supporting medical AI product R&D and medical device registration。Provides full datasets + custom annotation + incremental update services。
Data Compliance Statement
All images come from legally authorized sources; PHI fields are fully de-identified and contain no information that can directly identify an individual, in compliance with the Personal Information Protection Law and the Data Security Law。
Customized services
Supports extended requirements such as adding specific disease types, multimodal annotation expansion, and custom training of AI-assisted diagnostic models。
Quick Facts
AI Frontier Research
In 2025-2026, chest X-ray AI focused on lightweight models and XAI explainability. LXNet achieved nine-class classification and CD-DETR optimized small lesion detection, while multi-label classification and generative data augmentation became mainstream directions。