A machine learning model to classify aortic dissection patients in the early diagnosis phase

Area of Science:

• Cardiovascular Medicine
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Aortic dissection is a critical cardiovascular disease with high mortality rates.
• Timely diagnosis and treatment are essential but often delayed or missed.
• Misdiagnosis of aortic dissection leads to severe patient outcomes.

Purpose of the Study:

• To develop a data mining-based method for early classification and prediction of aortic dissection.
• To assist physicians in identifying potential aortic dissection cases among misdiagnosed patients.
• To improve diagnostic accuracy and reduce mortality associated with aortic dissection.

Main Methods:

• Utilized various machine learning algorithms for model development.
• Trained and tested models on a patient dataset using cross-validation.
• Focused on data mining techniques for classification and prediction.

Main Results:

• The Bayesian Network model demonstrated superior performance.
• Achieved a precision rate of 84.55% for predicting aortic dissection.
• Obtained an Area Under the Curve (AUC) value of 0.857.

Conclusions:

• The Bayesian Network model significantly aids physicians in the early diagnosis of aortic dissection.
• This approach can enhance clinical practice by improving diagnostic accuracy.
• Future research should incorporate diverse data for a more universal predictive model.