An Enhanced Feature Selection Integrated Explainable Machine Learning Models for Prediction of Breast Cancers

Abstract

The prediction of breast cancer has evolved significantly with the advent of machine learning models, which offer promising tools for early diagnosis and personalized treatment planning. This systematic literature review examines 31 peer-reviewed analyses undertaken between 2019 and 2024, targeting on the integration of enhanced feature selection methods within explainable machine learning models for breast cancer prediction. The primary goal is to assess how feature selection techniques contribute to model accuracy and interpretability, ensuring that predictions are both reliable and understandable to clinicians and researchers. The reviewed studies highlight a trend towards the use of sophisticated feature selection algorithms that refine input data, yielding models that not only enhance prediction accuracy but also generate actionable insights to inform decisions. This is especially important in healthcare, where the explainability of AI models can enhance trust and adoption by medical professionals, potentially improving patient outcomes. Despite these advancements, the review identifies several challenges, including the need for large, diverse datasets to ensure model generalizability and the difficulty of balancing model complexity with interpretability. Furthermore, the integration of these models into clinical workflows remains a significant hurdle due to varying levels of transparency and the potential for bias in feature selection. Gaps in the current literature suggest that future research concentrate on creating developing standardized frameworks for integrating feature selection with explainable models, ensuring that these tools are both effective and widely applicable in clinical settings. This review aims to guide future research and development efforts towards creating more robust, transparent, and clinically useful predictive models for breast cancer, ultimately contributing to more precise and personalized healthcare solutions.