Combining ensemble classification and integrated filter-evolutionary search for breast cancer diagnosis

Area of Science:

• Computational biology
• Medical informatics
• Machine learning applications in healthcare

Background:

• Breast cancer is a leading cause of death in women, making early diagnosis critical for survival.
• Technological advancements have spurred the development of intelligent medical assistants, particularly computerized diagnostic systems.
• Data mining and machine learning are increasingly vital in developing sophisticated diagnostic tools.

Purpose of the Study:

• To present a novel hybrid approach for breast cancer diagnosis.
• To improve the accuracy and efficiency of computerized diagnostic systems.
• To leverage data mining and machine learning for enhanced medical decision support.

Main Methods:

• A hybrid approach combining feature selection and classification using data mining techniques.
• Feature selection employs an integrated filter-evolutionary search with an evolutionary algorithm and information gain.
• An ensemble classification approach using neural networks with parameters optimized by an evolutionary algorithm.

Main Results:

• The proposed method was evaluated on real-world datasets from the UCI machine learning repository.
• Simulations demonstrated superior performance across various metrics including accuracy, precision, and recall.
• The method achieved an average improvement of 12% over existing state-of-the-art techniques.

Conclusions:

• The developed hybrid approach is effective for breast cancer diagnosis.
• The system functions as a capable intelligent medical assistant.
• The findings support the use of advanced computational methods in clinical diagnostics.