Medical Diagnosis based Bio-inspired Artificial Immune System

Abstract

Remedial disease diagnoses and timely provision of apt treatment always remained a challenge for healthcare professionals. Ongoing depletion of colossal healthcare data, can be converted to into valuable information, with reduced cost, wee processing time and improved diagnosis for life-threating diseases, such as heart diseases. As per World Health Organization (WHO), heart related diseases are the prime reason of mortality, representing 29% of deaths in Pakistan [1] and 31% of the global deaths [2]. On an average, 247 deaths per million, have been recorded and mostly due to delayed diagnosis in Pakistan. Being a major universal health concern, heart disease, was selected for medical diagnosis. Biological Human Immune System (BHIS) is a complex, robust and an adaptive system that defends human body from foreign or unknown pathogens; capable of self-monitoring, executing optimum disease detection, and active learning from data in memory cells. It comprises of group of molecules, which distinguish cells self (S) or non-self (NS), within the body. Several concepts from the BHIS are applied for real-world glitches using Artificial Immune System (AIS), which are ability to learn and memory to store. Bio-inspired techniques, are being employed in intellegent computing, security, and engineering. However, relatively little research focused on bio-inspired AIS based classifiers in machine learning for the prediction of the presence of heart disease. With headway in data science, benefits for healthcare industry are being realized across the globe. The key goal is to develop a diagnosis system for precisely envisaging existence of heart disease, through a novel Bio-Inspired Classification (BIC) based on Negative Selection Artificial Immune System (NSAIS), optimized through GA, with UCI Cleveland Dataset. Experimental outcomes show that bio-inspired algorithm outperformed six classical ML methods. This research not only aids in utilizing available medical data, but also enables the doctors to intelligently diagnose heart disease.