An Automatic Classification Method on Chronic Venous Insufficiency Images.

Area of Science:

• Medical imaging analysis
• Artificial intelligence in healthcare
• Vascular disease diagnostics

Background:

• Chronic venous insufficiency (CVI) impacts many, often lacking timely medical consultation.
• Accurate CVI severity classification is crucial for effective treatment planning.
• Physicians require efficient tools to assist in diagnosing CVI severity from medical images.

Purpose of the Study:

• To develop an automated classification method for assessing CVI severity.
• To create an AI-powered assistant tool for both patients and physicians.
• To improve the accuracy and efficiency of CVI diagnosis.

Main Methods:

• Utilized a concept classifier to map low-level image features to semantic features.
• Constructed a multi-scale semantic model for rich image representation.
• Employed a high-order dependency feature selection for training a scene classifier to estimate CVI severity.

Main Results:

• The CVI-classifier achieved high performance with 90.92% accuracy, 0.8735 kappa, and 0.9006 F1-score.
• The automated method demonstrated superior performance compared to human diagnosis based on images alone.
• Significant improvements in accuracy (9.11%), kappa (0.1250), and F1-score (0.0955) were observed over physician judgments.

Conclusions:

• The CVI-classifier offers a highly accurate and efficient solution for automated CVI severity classification.
• This AI tool can significantly aid healthcare professionals in diagnosing CVI.
• The proposed method holds promise for improving patient outcomes by enabling timely and precise medical interventions.