Abstract
Certain diseases and clinical conditions alter the exhaled breath gas components. Ammonia was identified as a biomarker for chronic kidney disease (CKD) from the laboratory analytical techniques. Although these techniques were accurate, lack of portability, time consumption and affordablity for the masses were the primary concern for the medical fraternity. To overcome these issues, a portable prototype was developed for the analysis of breath samples from patients undergoing hemodialysis. The aim and objectives of the study was as follows: (i) to detect ammonia in human breath sample using the commercial gas sensor (MQ135); (ii) to acquire breath patterns using NI-myDAQ for preprocessing and extraction of ten transient and steady state features; (iii) To classify the normal and abnormal breath signal using different neural network classifiers (Support Vector Machines, Naïve Bayes, and Multi-layer perceptron). Twenty-one subjects undergoing hemodialysis and 17 healthy subjects were included in the study. The obtained breath patterns revelaed that voltage flag runs between 0.6–0.9 V for a healthy individual, and 1.8–2.65 V for CKD patients. The variations in magnitude voltage were high for CKD subjects compared to healthy subjects. Among the obtained transient and steady-state features, normalized voltage change (ΔVm) showed a greater % difference between healthy and CKD patients undergoing haemodialysis. The multilayer perceptron achieved a better classification accuracy of 76.31% compared to SVM (60.52%) and naive bayes classifier (55.2%). Hence the developed prototype was simple in operation and cost effective, provides an affordable and act as complimentary diagnostic tool for CKD patients.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.