Applications of Machine Olfaction - The Electronic Nose Artificial Olfaction Technology - page 211

8.1.2.1 Bacteria Identification in Blood Culture Samples/Urine
Analysis/ENT Bacteria/Upper Respiratory
Discovery of diagnostic methods that give fast detection of the presence and type
of bacteria and so allowing proper antibiotic treatment can provide an increased
benefit to the patient as well as help reduce cost to the health care system. The use
of E-noses is to differentiate among various bacteria regularly found in blood
cultures [ 16 ], upper respiratory disease identification [ 17 ], ventilator-associated
pneumonia [ 18 ], and urine analysis [ 19 , 20 ].
The E-nose can distinguish among various common bacterial pathogens of the
upper respiratory area, including Staphylococcus aureus, Streptococcus pneumo-
niae, Haemophilus influenza, and Pseudomonas aeruginosa [ 17 ]. This is an
important application of electronic olfaction that could considerably improve the
current methodologies and be successfully used in clinical settings.
8.1.2.2 Ventilator Associated Pneumonia
Ventilator associated pneumonia (VAP) has a high rate of mortality and increased
incidence in critically ill individuals. Humphreys et al. explored the potential of
volatile fingerprints produced by different microorganisms not only in vitro but
also within clinical samples using an E-nose in perceptive between these groups.
They also tried to associate the microbiology culture results with the E-nose
responses for clinical samples [ 18 ]. The use of an E-nose is shown in recent studies
on tuberculosis for perception between various mycobacterial isolates in humans
[ 21 ].
8.1.2.3 Monitoring Haemodialysis
The study was done to analyze the potential for an ''E-nose'' as a basic monitoring
tool for haemodialysis. Blood samples were analyzed using an E-nose, which
comprised an array of 14 conducting polymer sensors, and these E-nose results
were compared to traditional biochemistry. E-nose demonstrated the ability to
distinguish predialysis blood from postdialysis blood independently, together with
an appropriate classification model, suitable for online monitoring [ 22 ]. The study
also suggests that E-nose technology might be a useful tool in discriminating
predialysis from postdialysis blood as well as control blood. Simultaneously, with
an appropriate classification model, it might be possible to build an online mon-
itoring system for the management of renal failure. Also it should be possible to
improve and modify currently used sensor arrays toward specific volatile markers
or marker groups, which would simplify the optimization of such applications.