Prediction Model for Detection of Chronic Kidney Disease: Logistic Regression

Abstract: Urinary System in human beings consists of two bean shaped elements called Kidneys. Purpose of urinary system is to eliminate waste, control levels of electrolytes, regulate blood pressure and blood volume, and control metabolism. Any malfunctioning of the elements of the urinary system may lead to imbalance of other connected systems of the body like circulatory system, digestive system and nervous system. Chronic Kidney Disease (CKD) is one ailment which could devastate the human body. It can be prevented via examining few indicators like RBC count, specific gravity value, Blood Pressure (BP), albumin levels in urine, sugar content, anaemia and WBC count. Other conditions like coronary artery disease, Diabetes Mellitus (DM) and bacterial infections could directly affect the kidneys. In this paper we have collected 400 samples from a public hospital and selected fields have been analysed for designing a prediction model for CKD. Logistic regression is carried out and accuracy, precision, and f1 score of the model has been measured. Various conclusions can be drawn from this interdependent data set and can be stored as historical data for future analysis. Keywords: Urinary System in human beings, Chronic Kidney Disease (CKD), RBC count, specific gravity value, Blood Pressure (BP), albumin levels in urine, sugar content, anaemia, WBC count, Logistic regression, accuracy, precision, and f1 score, coronary artery disease, Diabetes Mellitus (DM) and bacterial infections