CT Image Feature Diagnosis on the Basis of Deep Learning Algorithm for Preoperative Patients and Complications of...

Area of Science:

• Cardiovascular Imaging
• Artificial Intelligence in Medicine
• Medical Diagnostics

Background:

• Congenital aortic valve disease and severe aortic valve stenosis pose significant diagnostic challenges.
• Transcatheter aortic valve implantation (TAVI) is a critical intervention for these conditions.
• Accurate pre-procedural assessment is vital for successful TAVI outcomes.

Purpose of the Study:

• To evaluate the role of deep learning algorithms in analyzing CT angiography for congenital aortic valve disease and severe aortic valve stenosis.
• To assess the diagnostic accuracy of deep learning in measuring aortic valve dimensions.
• To determine the significance of deep learning in TAVI patient diagnosis and prognosis.

Main Methods:

• A cohort of 120 patients undergoing TAVI for aortic stenosis were included.
• CT angiography was performed within one week of diagnosis.
• Deep learning algorithms were employed to measure aortic valve dimensions using diameter, area, and perimeter methods.

Main Results:

• Deep learning demonstrated high accuracy in measuring aortic valve dimensions, with 95% confidence intervals for diameter (0.84-0.92), perimeter (0.68-0.87), and area (0.72-0.91).
• The study identified common comorbidities in TAVI patients, including hypertension (67%), diabetes (25%), and chronic renal insufficiency (11%).
• Deep learning showed excellent performance in CT image processing for TAVI patients.

Conclusions:

• Deep learning algorithms are highly effective in processing CT images for TAVI patients.
• This technology significantly aids in the diagnosis of aortic valve disease and associated complications.
• Utilizing deep learning in CT angiography analysis can improve TAVI treatment success rates and patient prognosis.