Feature Importance Score-Based Functional Link Artificial Neural Networks for Breast Cancer Classification.

Area of Science:

• Oncology
• Artificial Intelligence
• Machine Learning

Background:

• Breast cancer is a leading cause of death for women globally.
• Early diagnosis significantly improves treatment outcomes and survival rates.
• Existing machine learning methods for breast cancer classification can be computationally intensive and prone to overfitting.

Purpose of the Study:

• To propose a computationally efficient machine learning model for breast cancer classification.
• To address the overfitting problem in existing breast cancer diagnostic models.
• To leverage the Functional Link Artificial Neural Network (FLANN) for improved breast cancer diagnosis.

Main Methods:

• Utilized the Functional Link Artificial Neural Network (FLANN), a single-layer neural network.
• Employed the F-score metric to mitigate overfitting by selecting significant features.
• Tested the FLANN model on the Wisconsin Breast Cancer Dataset (WBCD) and Wisconsin Diagnostic Breast Cancer (WDBC) datasets.

Main Results:

• The proposed FLANN model demonstrated high performance in classifying breast cancer.
• Achieved an accuracy of 99.41% in breast cancer diagnosis.
• The F-score feature selection effectively reduced overfitting issues.

Conclusions:

• The FLANN model offers a promising and accurate approach for early breast cancer diagnosis.
• This method provides a computationally efficient alternative to complex machine learning techniques.
• The high accuracy suggests potential for real-world clinical application in breast cancer screening.