Académique Documents
Professionnel Documents
Culture Documents
Abstract—Patients with chronic obstructive pulmonary inflammation, airway hyperactivity, secretions and loss of
disease (COPD) are characterized by increased work of the structural integrity of the lung parenchyma [1].
breathing (WOB) and ventilatory muscle dysfunction.
Mechanical ventilation is applied to unload the WOB; rest A. Fuzzy Rule Based Systems and Neural Networks
respiratory muscles decrease arterial partial pressure of Fuzzy Rule Based Systems (FRBSs), constitute an
carbon dioxide (PaCO2) and treat hypoxemia. Since patients’ extension to classical rule based systems. Fuzzy Sets (FSs),
needs are not static, ventilator settings have to be adjusted were introduced by Lofti A. Zadeh in 1965 [2]. Knowledge
regularly. The aim of the present study was the development
representation is performed with the use of linguistic
and evaluation of a neuro-fuzzy controller, that utilizes non-
invasively acquired parameters for the determination of the
variables. Fuzzification is the process of interpreting a crisp
appropriate tidal volume (VT) and respiration frequency (RR) value as membership degree to the fuzzy sets which
ventilator settings for COPD patients. Forty three (43) hours of compose the fuzzy space. The total of linguistic terms and
non-invasively monitored physiology parameters and ventilator membership functions (µ) of a FRBS, forms the Data Base
settings, from four (4) different COPD patients ventilated in (DB) of the system. The logic of how the system responds to
control mode, were collected in two (2) General Hospitals in inputs, is formed as a collection of linguistic rules joined by
Greece. Recorded data were randomly allocated into two sets, the also operator. The total of rules forms the Rule Base
namely training set (60%) and evaluation set (40%). A neuro- (RB) of the system. In FRBSs such as Mandani [3], where a
fuzzy controller was developed and trained, by employing the
fuzzy output is produced, it is common to translate the fuzzy
training set. The controller utilizes non-invasively measured
parameters, namely oxygen saturation (SpO2), lung compliance
output to crisp values. This process is called defuzzification.
(C) and resistance (R), Peak Inspiratory pressure (PIP) and Fuzzy Logic (FL) demonstrates several advantages over
Plateau pressure (Pplateau), for predicting appropriate VT and other methodologies. It can easily model complex systems,
RR settings. The developed neuro-fuzzy controller was tested by introducing a development methodology similar to
against evaluation set. The Mean Square Error of the tidal human communication; experts’ knowledge is encoded
volume and the respiration rate was 0.222 ml/Kgr and 1.21 directly in a form very similar to their decision making
breaths per minute (bpm) respectively. process; the RB of a FRBS is evaluated in parallel, thus all
decision determinants are considered in the solution of a
I. INTRODUCTION problem; FL models uncertainty and imprecision in complex
3766
The proposed architecture utilizes NN to implement the data. Membership Functions (MFs) are automatically
rule base of a fuzzy system. Although both fuzzy and neural generated. The type of MFs is user defined, however the size
approaches possess remarkable properties when employed and position of the MFs is automatically adjusted to the
individually, their synergism is closer to human intelligence number of MFs and the fuzzy domain of each input-output
[6]. The employment of NN in fuzzy systems provides them variable. Excel input(s) and output(s) numeric values are
the ability to learn, while maintaining all the advantages of translated into membership degrees; input-output
fuzzy systems. membership degrees form the training set for the NN. The
NN presents results as membership degrees for the output
II. METHODOLOGY fuzzy domain(s). Finally NN output(s) membership degrees
Forty three (43) hours of non-invasively monitored are compared to Microsoft Excel ® output(s) membership
physiology parameters and ventilator settings, from four (4) degrees, for the NN performance evaluation.
different COPD ventilated in control mode, were collected FUN toolbox was developed for automatically generating
in two (2) general hospitals, namely Veteran’s General FRBSs with the synergism of NN, for a broad range of
Hospital of Athens (NIMTS), and University Hospital of applications, rather than being specific to the purpose of our
Heraclion, Crete (PEPAGNI). research.
The ICUs were equipped with Siemens-Draeger
ventilators, which are medibus serial interface enabled.
Furthermore the ICUs are equipped with a central
monitoring station, able to record patients’ parameters trends
for the duration of patient’s stay.
The ethics committees of the PEPAGNI and NIMITS
hospitals have approved the collection of patients;
physiology data. Patients’ data were collected directly from
digital outputs of medical equipment by using certified
medical software [13, 14].
We have designed and developed a MatLab
(Mathworks®) toolbox for automatically generating
Mandani FRBS [3] from available input(s) – output(s) data
in excel format.
The toolbox utilizes the NN toolbox of MatLab. The
toolbox was named FUzzy Neural (FUN) and allows for the
user to define fuzzy systems’ parameters as well as well as Fig. 4. Input-Output membership functions.
NN settings (Fig. 3). The structure of the toolbox is based on
Recorded data were randomly divided into training and
the architecture proposed by Wang et al [9].
evaluating sets. Sixty percent (60%) of the available data
were used for the training of the NN-FRBS system. The
controller utilizes non-invasively measured parameters,
namely oxygen saturation (SpO2), lung compliance (C) and
resistance (R), Peak Inspiratory pressure (PIP) and Plateau
pressure (Pplateau) as inputs for the system. The outputs are
suggested values for VT and RR ventilator settings. VT was
expressed as ml/Kgr, in order to allow comparison between
different patients.
Each input-output variable domain was assigned with five
Triangular- Trapezoid shape membership functions (Fig.
4).The defuzzification method chosen was the Center of
Area (Centroid).
The NN Architecture is graphically described in Fig. 5.
The applied NN consists of an input and an output layer and
two hidden layers. The number of input-output nodes is
Fig 3 A typical screen of the developed FUN Matlab (Mathworks ®)
defined by the number of fuzzy sets, while the hidden layers
toolbox for automatically generating FRBS from available input(s)-
output(s) recorded data. have 1000 and 100 nodes respectively. Tan-sigmoidal and
log-sigmoidal functions were chosen as appropriate transfer
The training process of the FUN toolbox is as follows: functions. Supervised learning of the NN is performed by
User selects the columns of the excel data that will be used comparing the membership degrees of the training sets
as input(s)-output(s) variables to the system, and the against the NN output.
corresponding fuzzy sets domain space. This feature allows
the user to develop FRBSs with a subset of the recorded
3767
First, the employment of the FUN toolbox for the
development of ventilation advisor systems, for patients
with lung pathologies other than COPD, such as Acute Lung
Injury (ALI) and Acute Respiratory Distress Syndrome
(ARDS). Second, the employment of the system for the
Fig. 5. NN architecture (Where IW(n,j) and LW(n,j) is the input (n) and
introductory training of ICU medical and nursing personnel.
layer weight respectively, b(j) is the bias and j is the neuron number).
Finally, the employment of the described tool-box for the
development of a decision supporting system, related to
III. EVALUATION AND RESULTS
home ventilation, since it is gradually accepted that is safer
The NN was trained for 1000 epochs, with the and less costly to treat stable, chronic ventilator-dependent
corresponding input-output membership degrees of the patients in their homes rather, than in an ICU bed, and that
training set values. The resulted NN, performed with a mean the patient in home-ventilation has a better quality of life,
square error (mse) of 0.0012. The mse refers to the deviation enhanced social relationships, and less likelihood of
from the actual membership degrees. nosocomial infections.
The developed NN-FRBS system was applied to the
evaluation set, in order to identify the deviation of the ACKNOWLEDGMENT
suggested ventilator settings from the recorded clinicians’
decisions. The mean square error of the tidal volume and the Authors would like to thank the ICU staff of the
respiration rate was 0.222 ml/Kgr and 1.21 bpm Veteran’s General Hospital of Athens (NIMTS), and the
respectively. Large deviations of suggested settings against University Hospital of Heraclion, Crete (PEPAGNI), for
recorded settings could be attributed to the failure of evenly their contribution to the presented research.
distributing acquired data into training and evaluation sets
due to random allocation. Fluctuations of the systems’ REFERENCES
output, suggests that the proposed system is responsive to [1] D.R.Hess, R.M.Kacmarek, “Essentials of mechanical ventilation”,
2002, 2nd edition, McGraw-Hill companies, ISBN 0-07-135229-5.
fluctuations of the input parameters. Figures 6a and 6b show [2] E.Cox, “The Fuzzy Systems Handbook”, 1994, AP Professional,
an instance of the suggested (solid) and the recorded (dash) ISBN 0-12-194270-8.
values of VT and RR. [3] E.H.Mandani, S.Assilian, “An experiment in linguistic synthesis with
a fuzzy logic controller”, 1975, International Journal of Man-Machine
Studies, vol 7, pp 1-13.
[4] P.Picton, “Neural Networks”, 2000, 2nd edition, Palgrave, ISBN 0-
333-80287-X.
[5] A.Tzavaras, P.R.Weller, B.Spyropoulos, “A Comparative Study of the
appropriateness of Neural Networks and Genetic Algorithms for the
Automatic Generation of Fuzzy Rule Based Systems from available
Training Data Sets”, 2007, 11th Int Conf on Cognitive & Neural Syst,
May 16-19, Boston Univ.
[6] L.H.Tsoukalas, R.E.Uhrig, “Fuzzy and Neural approaches in
engineering”, 1997, John Willey & Sons Inc., ISBN 0-471-16003-2.
[7] H.Tagaki, I.Hayashi, “NN-Driven fuzzy reasoning”, 1992,
International Journal of Approx. Reasoning, vol 5 (3), pp191-212.
[8] J.S.Jang, N.Gulley, “Fuzzy Logic toolbox for use with MATLAB”,
1995, The MathWorks Inc.
[9] X.Z. Wang, B.H. Chen, S.H. Yang, C. McGreavy, M.L. Lu, “Fuzzy
Rule Generation from Data for Process Operational Decision
Support”, 1997, Computational Chemical Engineering. 21 pp 661-
666.
Fig. 6. Top (6a) suggested (solid) vs. recorded (dash) VT. Bottom (6b)
[10] H.F.Kwok, D.A.Linkens, M.Mahfouf, G.H.Mills, “Rule-base
suggested (solid) vs. recorded (dash) RR.
derivation for intensive care ventilator control using ANFIS”, 2003,
Artificial Intelligence In Medicine, Vol 29, pp 185-201.
IV. CONCLUDING REMARKS [11] A.Wang, M.Mahfouf, G.H.Mills, “A blood gas hybrid model for
ventilated patients in ICU with new formulations for dead space and
The developed method and more specifically the outcome tidal volume”, 2006, Proc 24th IASTED Int. Multi-Conf. Biomedical
of the Neural- Fuzzy (FUN) MatLab (Mathworks®) toolbox Engineering., Innsburg, Austria (519-094 ACTA Press).
for automatically generating FRBS from available input(s) – [12] F Liu, G.S. Ng, C.Quek, T.F.Loh, “Artificial Ventilation Modeling
output(s) data in Microsoft Excel ® format, seems to give using Neuro-Fuzzy Hybrid System“, 2006, Int Joint Conf on Neural
Networks, Vancouver, BC, Canada, Jul 16-21, 2006, pp 5166-5171.
very good results after this first case study that was based on [13] Nortis: http://www.nortis.net
the data accumulated during forty three hours of non- [14] Siemens-Draeger: http://www.Draeger.com
invasively monitored physiology parameters and ventilator
settings, from four different COPD patients ventilated in
control mode.
Additionally, there are three more possible near-future
applications that are presently under investigation.
3768