Applications of Machine Olfaction - The Electronic Nose Artificial Olfaction Technology - page 220

Signal

Breath Gas

Signal processing
Element drawing out
Asthma classification

Gas sampling
bag

Sensor
Array

Gas collection

Sampling

Data analysis

Fig. 8.3 Proposed asthma analysis system in breath

increased concentration may be useful in distinguish asthma from other causes, i.e.,
chronic cough [ 44 ]. Dupont et al. presented that [16 ppb of NO in the lower respi-
ratory tract could be treated as a cutoff for asthma with a 90 % specificity and 90 %
positive predictive value. During research they analyze the diagnostic value of exhaled
NO capacity to differentiate between healthy persons with or without respiratory
symptoms and patients with confirmed asthma [ 45 ]. Thus it implies that noninvasive
measurement of exhaled NO is a good rival for asthma diagnosis.
This proposes a novel system that is special for breath analysis. We selected
chemical sensors that are sensitive to the biomarkers and compositions in human
breath, developed the system, and introduced the odor signal preprocessing and
classification method. To evaluate the system performance, we captured breath
samples from healthy persons and patients known to be afflicted with asthma and
conducted experiments on medical treatment evaluation and disease level identi-
fication. The breath sample is passed over the sensor and the sensor output is
stored in the PC through a software developed in LabVIEW as shown in Fig. 8.3 .
The sensor response of the patient is then compared with the database. Based on
the patient database developed, the treatment plan is suggested for the specific
level of asthma, if detected. If the sensor response for the sample is below the
reference line, the person is declared non-asthmatic. The reference line is fixed
through the breath sample of a healthy person.

Sensor Associated with Disease Condition

The chemical sensor capabilities can be identified with the system function and
performance. Each sensor in the array has a unique ''odour print,'' corresponding
to the compounds. The majority of the compounds are VOCs, but some are
inorganic compounds, such as ammonia, nitric oxide, carbon dioxide, and
hydrogen. Table 8.8 summarizes the main disease biomarkers and compositions in
human breath and the type of sensor required.

Sensor

In this work, we have utilized gas sensors, which are commercially available from
alpha sense ltd. and Figaro. These gas sensors have been commonly identified as