Automatic Diagnosis, Classification, and Segmentation of Abdominal Aortic Aneurysm and Dissection from Computed...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Cardiovascular Disease

Background:

• Abdominal aortic aneurysm (AAA) and abdominal aortic dissection (AAD) are critical cardiovascular diseases with fatal implications.
• Accurate and efficient diagnosis of AAA and AAD is strategically important for patient outcomes.
• Current diagnostic methods can be time-consuming and require specialized expertise.

Purpose of the Study:

• To develop and evaluate a novel deep learning-based approach for the automated diagnosis of AAA and AAD from CT images.
• To achieve accurate detection and segmentation of AAA and AAD regions.
• To improve the efficiency and reduce the workload associated with diagnosing these conditions.

Main Methods:

• A hybrid deep learning model combining a convolutional neural network (CNN) and a pyramid scene parsing network was developed.
• The architecture effectively extracts features from CT scans for classification and delineation of diseased regions.
• The model was evaluated using Python programming to assess its accuracy and performance metrics.

Main Results:

• The proposed deep learning strategy achieved an average accuracy of 89.64% in diagnosing AAA and AAD.
• The model demonstrated a high intersection over union (IoU) of 83.76% for segmentation accuracy.
• Performance metrics surpassed existing methods like ResDenseUNet, INet, and C-Net.

Conclusions:

• The developed deep learning strategy shows significant promise for the automatic diagnosis of AAA and AAD.
• This automated approach can effectively reduce the diagnostic workload for cardiovascular surgeons.
• The technique offers a potential advancement in the clinical management of aortic diseases.