A survival classification method for hepatocellular carcinoma patients with chaotic Darcy optimization method based...

Area of Science:

• Medical Informatics
• Data Science
• Bioinformatics

Background:

• Surveys are vital for data retrieval but often suffer from missing data and redundant features, hindering effective knowledge extraction.
• Addressing missing data and feature selection are critical steps for improving the utility of survey data in research.

Purpose of the Study:

• To develop and evaluate a novel classification method for Hepatocellular Carcinoma (HCC) survival prediction.
• To address challenges of missing data and feature redundancy in survey datasets.

Main Methods:

• A supervised feature completion method using statistical moments (average) was applied to the HCC survey data.
• A chaotic Darcy optimization algorithm was employed for feature selection, identifying the most discriminative features.
• The combined approach was used for HCC survival classification.

Main Results:

• The proposed method successfully completed missing features in the HCC survey dataset.
• Feature selection identified 31 discriminative features from the completed dataset.
• The chaotic Darcy optimization-based HCC survival classification achieved a high accuracy rate of 0.9879.

Conclusions:

• The integrated approach of average-based feature completion and chaotic Darcy optimization-based feature selection is effective for knowledge extraction from survey data.
• This method significantly improves the accuracy of Hepatocellular Carcinoma survival classification.
• The study highlights the potential of swarm optimization techniques in medical data analysis.