Neuro-fuzzy icu ventilator patients modeling

Article type: Research Article

Authors: Wahab, Abdul^{; *} | Chai, Quek | A H, Luah

Affiliations: Center for Computational Intelligent, School of Computer Engineering Nanyang Technological University, Blk N4 #2A-36, Nanyang Avenue, Singapore 639798

Correspondence: [*] Corresponding author. Dr. Abdul Wahab, Tel./Fax: +65 6790 4948/+65 6792 6559; E-mail: [email protected].

Abstract: Artificial ventilation is an important treatment for the patients in the intensive care unit (ICU). An important step in building an efficient decision support system for intelligent ventilation is to develop a patient model. The patient model gives an idea of how the patient will behave under different ventilator settings. The feasibility of using neuro-fuzzy systems to model the physiology of the ventilated patient was assessed. Feature selection was also performed to explore other important input variables that could model the physiological state of the patients more accurately. Even though the overall patient model do not provide a good matched prediction of the patient's arterial oxygen tension, but it is still able to provide satisfactory arterial carbon dioxide tension predictions. This thus opens up new opportunity to model the ventilated patient.

Keywords: ECG, MFCC, GMM, MLP, ANFIS, RBF, GRBF, ANFIS, EFuNN

DOI: 10.3233/JCM-2009-0244

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 9, no. s2, pp. S159-S167, 2009