Anemia prediction using gene expression programming (GEP) and explainable artificial intelligence approaches.

Area of Science:

• Computational Biology
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Anemia is a widespread global health issue, disproportionately affecting vulnerable populations.
• Traditional anemia diagnosis methods are often slow, expensive, and prone to human error.
• Developing efficient and accurate diagnostic tools is crucial for managing anemia's adverse effects.

Purpose of the Study:

• To investigate the efficacy of Gene Expression Programming (GEP), a machine learning technique, for predicting anemia.
• To optimize GEP model hyperparameters for maximum predictive accuracy.
• To enhance model interpretability using Explainable AI (XAI) methods.

Main Methods:

• Utilized a publicly available Kaggle dataset containing clinical parameters like hemoglobin and red blood cell indices.
• Applied Gene Expression Programming (GEP) for anemia classification.
• Employed Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive exPlanations (SHAP) for model interpretability.

Main Results:

• The optimized GEP model achieved a high accuracy rate of 99.30% in predicting anemia.
• Hemoglobin levels and gender were identified as the most significant predictive features.
• XAI methods provided insights into the decision-making process of the GEP model.

Conclusions:

• Machine learning, specifically GEP, offers a powerful and accurate approach for anemia diagnosis.
• The developed model is dependable, automatic, and interpretable, demonstrating the potential of ML in clinical settings.
• This research highlights the value of integrating advanced computational methods into medical diagnostics for improved patient outcomes.