Neural Network For The Estimation Of Ammonia Concentration In Breath Of Kidney Dialysis Patients

Abstract

Neural networks are an extremely powerful tool for data mining. They are especially useful in cases involving data classification where it is difficult to establish a pattern in the search space. In an era when artificial intelligence is increasingly being utilised in industrial and medical applications throughout the world, it is becoming evident that this is an emerging trend. This paper explores the idea of artificial intelligence by employing the use of a feed-forward neural network with two process layers to determine the concentration of ammonia in exhaled human breath. The human mouth contains many kinds of substances both in liquid and gaseous form. The individual concentrations of each of these substances could provide useful insight to the health condition of the entire body. Ammonia is one of such substances whose concentration in the mouth has revealed the presence or absence of diseases in the body. Kidney failure is one diesease which is identified by an extremely high ammonia content in human breath. This disease is as a result of the kidneys’ inability to process the body’s liquid waste. The result of this is the release of urea throughout the body which is dissipated in the form of ammonia through oral breath. The neural simulation is carried out using NeuroSolutions version 5 software. The neural network correctly identified the concentration of oral ammonia as an indication of kidney failure with an accuracy of 85%.