Artificial Intelligence Models for Predicting Iron Deficiency Anemia and Iron Serum Level based on Accessible Laboratory Data

Abstract

Background: One of the most important strategies for treating anaemia is early detection and therapeutic intervention to avoid irreversible organ damage. The aim of this study is to demonstrate artificial intelligence models for predicting iron deficiency anaemia and iron serum levels using available laboratory data.

Methodology: Three AI models, Naïve Bayes, Support vector machine (SVM) and Convolutional neural network (CNN) were developed to detect anaemia in captured and processed images (datasets).

Results: The CNN achieved the highest accuracy of 90.27%, while SVM had the lowest accuracy at 64.66%. Naïve Bayes achieved 89.96% accuracy.This demonstrates that data mining software is efficient and effective at detecting diseases such as iron deficiency anaemia and deficiency of iron serum levels.

Our findings illustrated that, using machine learning and explainable AI to detect anaemic conditions in blood has the potential to improve patient outcomes and benefit the healthcare industry. After being trained, validated, and tested on the datasets, the proposed models produced a significant outcome. This demonstrates that data mining software is efficient and effective at detecting diseases such as iron deficiency anaemia and deficiency of iron serum levels.

Conclusions: By creating a predictive AI model that shows promise for improving diagnostic precision and clinical decision-making in anaemia management. By leveraging data-driven approaches and advanced computational tools, healthcare systems can improve patient care, lower healthcare costs, and improve overall public health outcomes. Our study demonstrated that the use of non-invasive approaches, such as machine learning algorithms, is effective, less costly, takes less time, and produces excellent results for anaemia detection