Enhancing Disease Diagnosis: Statistical Analysis of Haematological Parameters in Sickle Cell Patients, Integrating Predictive Analytics

Abstract

Sickle cell disease (SCD) affects 30 million people worldwide, causing a range of symptoms from mild to severe, including Vaso occlusive crises (VOC). SCD leads to damaging cycles of sickling and desickling of red blood cells due to HbS polymer formation, resulting in chronic haemolytic anaemia and tissue hypoxia. We propose using machine learning to categorize SCD patients based on haemoglobin, reticulocyte count, and LDH levels, crucial markers of hemolysis. Statistical analysis, particularly Linear Regression, demonstrates how haemoglobin depletion occurs using LDH and reticulocyte parameters. Bilirubin and haemoglobin, two integral biomarkers in clinical biochemistry and haematology, serve distinct yet interconnected roles in human physiology. Bilirubin, a product of heme degradation, is a critical indicator of liver function and various hepatic disorders, while haemoglobin, found in red blood cells, is responsible for oxygen transport throughout the body. Understanding the statistical relationship between these biomarkers has far-reaching clinical implications, enabling improved diagnosis, prognosis, and patient care. This research paper conducts a comprehensive statistical analysis of bilirubin and haemoglobin using various regression techniques to elucidate their intricate association. The primary objective of this study is to characterize the relationship between bilirubin and haemoglobin. Through meticulous data analysis, we explore whether these biomarkers exhibit positive, negative, or no correlation. Additionally, this research develops predictive models for estimating haemoglobin levels based on bilirubin data, offering valuable tools for healthcare professionals in clinical practice.