Machine Learning Approach for Classification of Sickle Cell Anemia in Teenagers Based on Bayesian Network

Abstract

In this study, we employed a Bayesian network approach for the classification of sickle cell anemia in teenagers based on their medical data. Sickle cell anemia is a hereditary blood disorder characterized by the presence of abnormal hemoglobin, leading to distorted red blood cells. Early detection and classification of this condition are crucial for timely intervention and improved patient outcomes. Our research focused on leveraging the algorithmic power of Bayesian network to model and analyze a diverse set of medical parameters in teenagers, including age, platelet count, mean corpuscular hemoglobin concentration (MCHC), red blood cell count, packed cell volume etc. The Bayesian network method employed for the classification of sickle cell anemia involves using a probabilistic graphical model to represent the relationships among different medical parameters. The model incorporates Bayesian principles to update and refine its predictions as new information is introduced. The method identifies key features in the dataset that contribute significantly to the classification, providing valuable insights for early detection and intervention. The Bayesian network demonstrated remarkable efficacy in accurately classifying teenagers as either positive for sickle cell anemia or negative, achieving an impressive 99% accuracy rate. This high level of accuracy indicates the robustness of the model in discerning intricate patterns within the medical data. Key features contributing to the classification are found in the dataset, shedding light on their relevance in distinguishing between positive and negative cases of sickle cell anemia, especially in teenagers. Our findings provide valuable insights into the potential diagnostic significance of sickle cell anemia classification in the teenage population. This research contributes to the growing body of knowledge in the field of medical informatics and computational biology, offering an efficient and reliable tool for healthcare practitioners in the early identification of sickle cell anemia in teenagers. The demonstrated accuracy of the Bayesian network shows its potential as an effective decision support system, aiding clinicians in making informed decisions and facilitating timely interventions for improved patient care.

Keywords

• sickle cell, anemia, teenagers, bayesian network, machine learning