Identifying Acute Aortic Syndrome and Thoracic Aortic Aneurysm from Chest Radiography in the Emergency Department...

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Cardiovascular Diagnostics

Background:

• Timely diagnosis of acute aortic syndrome (AAS) and thoracic aortic aneurysm (TAA) is critical in emergency departments (EDs) due to their high mortality.
• Chest pain and back pain are common chief complaints in the ED, often necessitating differentiation between life-threatening conditions and less severe causes.

Purpose of the Study:

• To evaluate the efficacy of convolutional neural network (CNN) models in analyzing chest radiography (CXRs) for the detection of AAS and TAA.
• To explore the potential of AI-driven imaging analysis to enhance diagnostic accuracy and efficiency in emergency settings.

Main Methods:

• A retrospective case-control study involving 1625 adult patients presenting to the ED with chest or back pain.
• Chest radiography (CXR) data were split into training (80%) and testing (20%) sets.
• Four different CNN models were trained on the CXR data to identify AAS and TAA.

Main Results:

• The InceptionV3 CNN model demonstrated the highest performance with an F1 score of 0.76.
• The study successfully trained CNN models to analyze CXRs for suspected AAS and TAA.

Conclusions:

• CNN-based analysis of CXRs offers a novel tool for clinicians managing patients with suspected AAS and TAA in the ED.
• Integrating such AI-powered imaging tools into the ED workflow could significantly improve the diagnosis of critical aortic diseases.