Premium
Advances in multisensor information fusion: A Markov–Kalman viscosity fuzzy statistical predictor for analysis of oxygen flow, diffusion, speed, temperature, and time metrics in CPAP
Author(s) -
Rodger James A.
Publication year - 2018
Publication title -
expert systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 38
eISSN - 1468-0394
pISSN - 0266-4720
DOI - 10.1111/exsy.12270
Subject(s) - computer science , kalman filter , continuous positive airway pressure , fuzzy logic , markov chain , noise (video) , diffusion , simulation , artificial intelligence , machine learning , medicine , obstructive sleep apnea , thermodynamics , anesthesia , physics , image (mathematics)
The efficacies of continuous positive airway pressure (CPAP) are well documented in decreasing the apnoea–hypopnoea index in patients with obstructive sleep apnoea. To guarantee these efficacies, CPAP manufacturers must thoroughly test these devices to ensure the flow of oxygenated air to the patient at various temperatures during a prescribed time frame. The calculation of the percent oxygen in a “bimixture” of gas can be done by measuring the travel time of a sound wave through the gas, and the travel time is proportional to the density of the air. We utilized existing multisensor tubes that were developed to collect and measure oxygen flow, diffusion, speed, temperature, and time metrics. Then these metrics were analysed using a Markov fuzzy, statistical, artificial neural network, nearest‐neighbour predictive approach to determine the interactions between these variables. An improved Kalman filter method was employed to reduce noise, increase viscosity, and obtain correct data from the CPAP system.