Detection of suspected lung disease areas from chest X-ray images
Displays the location of abnormal findings such as Consolidation, Pneumothorax, Fibrosis, Lung Nodule/Mass, and Pleural Effusion.
AI analysis results are visualized in three types (Heatmap, Grayscale Contour, and Combined) according to user settings to provide intuitive images.
DC-XR-03
Approval from the Ministry of Food and Drug Safety (No. 21-841)
Consolidation (CSN)
Sensitivity
87%
Specificity
93.6%
* source: SMG-SNU Boramae Medical Center, Seoul·Pusan National University Hospital, Busan
Pneumothorax (PTX)
Sensitivity
Specificity
94.4%
99.63%
* source: SMG-SNU Boramae Medical Center, Seoul·Pusan National University Hospital, Busan
Fibrosis (FIB)
Sensitivity
Specificity
88.10%
99.56%
* source: Chung-Ang University Hospital, Seoul
Nodule·Mass (NDL)
Sensitivity
Specificity
83.41%
94.95%
* source: Chung-Ang University Hospital, Seoul
Pleural Effusion (PEF)
Sensitivity
Specificity
87.08%
99.58%
*source: Chung-Ang University Hospital, Seoul
1. AI-related medical images for each disease
It analyzes data and aids in reading by medical institutions and doctors.
2. Quickly locate the disease
Detects and alerts you to anything suspicious.
3. Easy PACS interworking / It is interlocked with the PACS in the hospital through simple DICOM communication.
+479,197 datas
Features
Rich Learning Data
High accruacy of diagnosis is achieved with approximately 479K global dataset.
Analysis Report
DEEP:CHEST provides a image analysis report with evaluation result, contributing to an enhanced medical workflow environment.
Key Data Provided
• Number of detected abnormalities
• Highest probability value among detected abnormalities
• Highest probability value by indication among detected abnormalities
• Types of detected indications
AI-informed Treatment Decisions in Minutes
Changes that DEEP:CHEST will bring
Accuracy
Increase diagnostic accuracy by marking areas suspected of lung disease on chest X-ray images with outlines and probability values.
Efficiency
It reduces medical staff's interpretation time and improves diagnostic efficiency.
Point of Care
Focus on critically ill patients and support rapid decision-making in emergency situations with abnormal probability information.
Publications
-
Ro Woon Lee, et al., “Comparative Analysis of M4CXR, an LLM-Based Chest X-Ray Report Generation Model, and ChatGPT in Radiological Interpretation”, Journal of Clinical Medicine, 2024.
-
Sun Yeop Lee, et al., “Localization-Adjusted Diagnostic Performance and Assistance Effect of a Computer-Aided Detection System for Pneumothorax and Consolidation”, Npj Digital Medicine, 2022.
