Volume 15 Supplement 3
Neuro-fuzzy classification of asthma and chronic obstructive pulmonary disease
© Badnjevic et al.; 2015
Published: 11 September 2015
This paper presents a system for classification of asthma and chronic obstructive pulmonary disease (COPD) based on fuzzy rules and the trained neural network.
Fuzzy rules and neural network parameters are defined according to Global Initiative for Asthma (GINA) and Global Initiative for chronic Obstructive Lung Disease (GOLD) guidelines. For neural network training more than one thousand medical reports obtained from database of the company CareFusion were used. Afterwards the system was validated on 455 patients by physicians from the Clinical Centre University of Sarajevo.
Out of 170 patients with asthma, 99.41% of patients were correctly classified. In addition, 99.19% of the 248 COPD patients were correctly classified. The system was 100% successful on 37 patients with normal lung function. Sensitivity of 99.28% and specificity of 100% in asthma and COPD classification were obtained.
Our neuro-fuzzy system for classification of asthma and COPD uses a combination of spirometry and Impulse Oscillometry System (IOS) test results, which in the very beginning enables more accurate classification.
Additionally, using bronchodilatation and bronhoprovocation tests we get a complete patient's dynamic assessment, as opposed to the solution that provides a static assessment of the patient.
Asthma is an inflammatory impairment of airways, which as result becomes hyperreactive and generates increased mucus, mucosal swelling and contraction of smooth airway muscles, all of which contribute to airway obstruction. In addition, the prevalence of asthma has been increasing in the recent decades, as well as the costs for asthma care [1–3]. One of the problems related to asthma management is that a number of patients with asthma are either improperly diagnosed or misdiagnosed for having other respiratory diseases such as a common cold, acute bronchitis or chronic obstructive pulmonary disease (COPD) [4, 5].
COPD is a respiratory disorder characterized by chronic and recurrent airflow obstruction, which increases resistance and dynamic hyperinflation of the peripheral airways [6, 7]. The two main examples of this are obstructive emphysema and chronic bronchitis. Approximately 75% of all COPD patients do not have diagnosis. Most of them have a mild degree COPD, but among them a 4% are with severe degree and 1% with very severe degree of COPD . Approximately 200,000 to 300,000 people in Europe die every year due to COPD, and the number exceeds the quote of lung cancer and breast cancer together [9–11].
Based on presented diagnosis problems there is a need for the system which can help physicians to make more successful diagnosis.
The most commonly used pulmonary function tests that are used to detect asthma and COPD are spirometry and Impulse Oscillometry System (IOS) [12, 13]. Neuro-fuzzy systems are used for the detection different types of diseases. In 2004, Barua et al  presented a system based on trained neural networks that uses the results of measurements performed by IOS for classification of asthma. In 2009, Winkler et al  were able to diagnose asthma and COPD patients using different measurement methods on IOS. Asaithambi et al  in 2012 made classification of respiratory abnormalities using adaptive neuro fuzzy inference system based only on spirometric measurements. As can be seen from the papers [14–16] the authors have tried to using their neuro fuzzy systems to perform diagnosis of disease based on the results of lung function measurements using only IOS or spirometry. In 2008, Meraz et al  designed software for classification of respiratory disease in children. The same year Hafezi  made an integrated software package for model-based neuro-fuzzy classification of small airway dysfunction, based on augmented RIC (aRIC) and extended RIC (eRIC) equivalent lung models and IOS data.
The most successful diagnosis is achieved in combination of IOS and spirometry measuring test results [6–9]. Based on it a static assessment of the patient is obtained, which is achieved only on symptoms and first pulmonary function testing, without any further monitoring of patients. In order to get dynamic assessment of the patient, it is necessary to take the patient's symptoms and allergies in consideration, to perform auscultation of the patient, and to apply bronchial dilation (BDT) and bronchial provocation tests (BPT). After BDT and/or BPT treatment and after the second and/or third measurement of lung function, potential changes of pulmonary parameter values may be presented, from which physicians get accurate information on the specifics of the disease. Dynamic assessment of the patient is necessary process according to Global Initiative for Asthma (GINA) and Global Initiative for chronic Obstructive Lung Disease (GOLD) guidelines [19, 20].
This paper presents a neuro-fuzzy system for classification of asthma and COPD based on GINA and GOLD guidelines.
Neuro-fuzzy system design
Input data for our fuzzy rules are results from .pdf files obtained from spirometry (SPIR) and IOS measurement reports. Those data are used to preliminary perform classification of diseases, and they present input vector of the neural network. Symptoms based on GINA and GOLD guidelines and ANN classification results gives proposed diagnosis.
Asthma was defined based on history of recurrent wheezing, cough, chest tightness or shortness of breath, in accordance to allergic symptoms, which could have a seasonal pattern or might exacerbate at night or by common exposure to allergen, exercise or smoking. Patients with positive symptoms and Resistance at 5 Hz (R5)>150%pred, Resistance at 20 Hz (R20)>150%pred, Reactance at 5 Hz (X5pred - X5)>0.15 kPa/(L/s), Ratio of Forced Expiratory Volume in 1 second and Forced Vital Capacity (FEV1/FVC)<0.8 were diagnosed with asthma. Patients with nonspecific test results were dedicated for bronchial hyper reactivity (BHR), first with BDT, and then with BPT. Patients are diagnosed with asthma when their R5 decreases by more than 25% (ΔR5), the absolute value of X5 decreases by more than 20% (ΔX5) and FEV1 decreases by more than 12% of baseline value and 200 ml. In cases where BDT did not achieve any improvement for final diagnose, patients were examined with methacholine BPT. The doses of methacholine, serially increased, were administered, and FEV1 was measured after each methacholine dose until a 20% decrease in FEV1. After this procedure, the system was able to make the final classification of asthma.
COPD was diagnosed based on positive history of dyspnea (progressive, exertion or persistent), chronic cough (intermittent or non-productive), chronic sputum and a history of exposure to tobacco, occupational dusts, chemicals or other smokes at an age above 40. The diagnosis of COPD was confirmed by having a positive history, R5>150%pred, (X5pred - X5)>0.15 and FEV1/FVC<0.7. Patients with nonspecific test results were evaluated for bronchial hyperreactivity (BHR) by means of BDT, in order to check which group of COPD they belonged to (A, B, C or D) . The bronchial dilation test was performed according to international guidelines using 400 μg of salbutamol. After the completion of the test, the system was able to make the final classification of COPD.
Fuzzy logic design
where: μnormal(R5) is membership function of reference values of R5 from IOS, μhigh(R5) is membership function of upper limit of normal (ULN) values of R5 from IOS, μnormal(X5) is membership function of reference values of X5 from IOS, μhigh(X5) is membership function of upper limit of normal values of X5 from IOS, μnone(ΔR) membership function for none difference between R5 and R20, μmin(ΔR) is membership function for minimum difference between R5 and R20 and μsig(ΔR) is membership function for significant difference between R5 and R20. Fuzzy values for R20 are the same as for R5. Input values can be calculated by using fuzzy T-norm operator .
if (R5 is normal) and (R20 is normal) and (R5-R20 is none) and (X5 is normal) then (output1 is NSCO-SV);
if (R5 is normal) and (R20 is normal) and (R5-R20 is minimum) and (X5 is normal) then (output1 is NSCO-DV);
if (R5 is normal) and (R20 is normal) and (R5-R20 is significant) and (X5 is normal) then (output1 is NSPO);
if (R5 is ULN) and (R20 is ULN) and (R5-R20 is none) and (X5 is normal) then (output1 is CO-SV);
if (R5 is ULN) and (R20 is ULN) and (R5-R20 is minimum) or (X5 is normal) or (X5 is ULN) then (output1 is CO-DV);
if (R5 is ULN) and (R20 is ULN) and (R5-R20 is significant) and (X5 is ULN) then (output1 is PO);
where NSCO-SV is Negative Sub Central Obstruction - Same Values, NSCO-DV is Negative Sub Central Obstruction - Different Values, NSPO is Negative Sub Peripheral Obstruction, CO-SV is Central Obstruction - Same Values, CO-DV is Central Obstruction - Different Values, PO is Peripheral Obstruction, while output1 is suggestion of classification based on IOS test report.
Artificial neural network design
The linear feed forward (FF) neural network is, according to the application experts, sufficient to properly perform the classification and these types of NN are mostly used for classification task [26–30]. The network is divided into layers. The input layer consists only of network inputs. It is then followed by a hidden layer which consists of a number of neurons, or hidden units placed in parallel. Each neuron performs a weighted summation of the inputs, which is then passed to some non-linear activation function σ, also called the neuron function.
In our case, we choose tansig activation function for hidden layer which is equivalent to the hyperbolic tangent function. This function is used in highly non-linear data classification  which is true in our case.
The network output is formed by another weighted summation of the outputs of the neurons in the hidden layer. This summation on the output is called the output layer. Generally, the number of output neurons equals the number of outputs of the approximation problem. We choose linear activation function for the output layer, because that is commonly used in regression problems .
Artificial neural network (ANN) is trained using estimation data and validated using validation data. We used 80-20 division to get estimation and validation data from the data set. The training data set contains 1000 tests previously obtained from database of the company CareFusion, 800 of them were used for estimation, and other 200 for validation. The training algorithm we used is Levenberg-Marquardt (LMA), which is common training algorithm in data classification .
Neuro-fuzzy system validation
Basic information about patients involved in this study.
19.85 ± 8.185
52.25 ± 7.636
30.03 ± 11.833
Before we began with our study and research we obtained ethics board approval for human subject from Clinical Centre University of Sarajevo. All the volunteers gave written informed consent.
For purpose of this study, baseline assessment for all patients was to answer on questions regarding symptoms, allergies, history and risk factors of asthma and COPD according to GINA and GOLD guidelines to physicians.
In another step, lung function was measured incorporating spirometry and IOS test results. All pulmonary function test reports were obtained using the CareFusion Germany "Master Screen IOS" device, which allows performing spirometric and impulse oscillometric lung function tests.
Using spirometry, forced vital capacity (FVC) and forced expiratory volume in one second FEV1 were derived, while the ratio FEV1/FVC was calculated. The impulse oscillometry system (IOS) provided respiratory resistance R5 at 5 Hz, proximal resistance R20 at 20 Hz, lung reactance X5 at 5 Hz and resonant frequency Fres [34, 35]. All measured results were compared to predicted (pred) values of the patient.
If the patients are with nonspecific test results then it is necessary to do BDT and/or BPT. The bronchial dilation test was performed according to international guidelines [36, 37] using 400 μg of salbutamol. In cases where BDT did not achieve any improvement for final diagnose, patients were examined with methacholine BPT. After this procedure, the neuro-fuzzy system will make the final classification of asthma, COPD or normal lung function.
ANN training results
Estimation and validation data for NN.
Number of reports
Estimation set size
Validation set size
No. of reports
% of true diagnosis
No. of reports
% of true diagnosis
Comparison between Feed Forward and Layer Recurrent ANN.
Time to train with LMA [s]
Peak memory consumption [MB]
Feed Forward ANN
Layer Recurrent ANN
System validation results
Effectiveness of the neuro-fuzzy system in classification of asthma, COPD and healthy subjects.
From total sum of 170 asthmatic patients the software correctly classified 149 patients after IOS and spirometry test results, 17 patients after BDT results, 4 patients after BPT results, while incorrectly classified 1 patient after BDT results.
From total sum of 248 COPD patients the software correctly classified 212 patients after IOS and spirometry test results, 34 patients after BDT results, while incorrectly classified 2 patients after BDT results.
It also may be noted that the software classified majority of the patients (over 85%) instantly upon the completion of standard tests (spirometry and IOS), without a single failure.
Confusion matrix for performance of implemented algorithm.
Test outcome positive
Positive predictive value
False discovery rate
Test outcome negative
False omission rate
Negative predictive value
True positive rate
False positive rate
False negative rate
True negative rate
The best and final results are achieved only after completing the entire process of classification of disease, i.e. after obtaining a complete dynamic assessment of the patient.
When making diagnosis of a respiratory disease, physicians use results obtained from spirometry and IOS. These results show that there is a correlation between parameters obtained by spirometry and IOS. In 2011, the Mehdi Nikkhah et al  compared methods of spirometry and IOS measurements in diagnosis of asthma and COPD patients. Furthermore, they demonstrated a correlation between IOS parameters and FEV1 parameter in asthmatics. They had shown that only R5 had a correlation with FEV1 in COPD patients, and that the progress of COPD disease is closely associated with the R5 parameter. Also, they had shown that in the case of their subjects sensitivity of X5 parameter in COPD patients is 76%, whereas in asthmatics sensitivity of R20 is 77%. In their studies, Kolsum et al in 2008  and Song et al in 2009  had already shown the correlation between R5 and X5 with FEV1. Since studies have shown that the combination of spirometry and IOS measurements achieve more successful diagnosis, the method which we developed for classification of asthma and COPD is using both type of measurements.
In 2009, Winkler et al  were able to diagnose 87-94% of asthma and COPD patients only when using different measurement methods on IOS. The same year Chakraborty et al  presented an intelligent diagnostic system for bronchial asthma based on symptoms and questions to which patients responded and obtained 90.03% accuracy in diagnosing asthma. In 2004, M. Barua et al  presented a system based on trained neural networks that uses the results of measurements performed by IOS for classification of asthma. They tested the system on 131 patients and obtained a 98% exact classification with known patterns, while their neural network, when working with unknown patterns, confirmed exact classification in 61% of cases. In our study, from a total sum of first time entered patients, after using only standard tests, the correct classification of patients was achieved in 87.65% of cases of asthma and 85.50% of COPD. If we use BDT and/or BPT suggested by neuro-fuzzy system, then we achieve correct classification in 99.41% of cases of asthma and 99.19% of COPD patients.
In 2006, Price D.B. et al  developed a tool in the form of a quiz based on the questions in order to recognize COPD in smokers and suggested further testing, while in the same year Tinkelman DG et al  further expanded their diagnostic tool for asthma and COPD. They tested it on 597 patients and obtained a sensitivity of 72.0% and specificity of 82.7%. While testing developed neuro-fuzzy system on our 455 study patients, we obtained sensitivity of 99.28% and specificity of 100%.
Tinkelman DG et al also presented the degrees in which the diagnosis of asthma and COPD are usually established, in the same way as presented in Figure 3 of our study.
In 2008, E. Meraz et al  and in 2009 N. Hafezi et al , based on known equivalent electrical models of lungs and their values specified for the healthy and diseased patients [42–49], developed a computational tool that classifies respiratory diseases in children by using IOS results. The advantage of our previous research [50–52] and new solution with respect to the one presented by E. Meraz et al  and N. Hafezi et al  is that our solution uses a combination of spirometry and IOS classification test results, which in the very beginning enables more accurate classification. Also, for the classification of diseases, in addition to the results obtained by using spirometry and IOS, symptoms according to GINA and GOLD rules as well as bronchiodilatory tests are necessary for proper classification of asthma and COPD. In this way, we get a complete patient's dynamic assessment, as opposed to the solution that provides a static assessment of the patient. The solution presented by Meraz  and Hafezi  is based on the equivalent electrical models of lungs, and those values are obtained from previous studies where age and race of the patients have to be taken into account, which is not the case in our solution.
In 2007, G. Coppin et al  introduced a computer system based on neural networks that detects emphysema using digital x-rays performed on patients with COPD, where they had accuracy of 90% on 161 subjects. This can be a great tool in combination with neuro-fuzzy system presented in our study.
In the future work, the aim of the researchers is to develop rules of fuzzy logic and neural network training for other respiratory diseases that can be determined on the basis of lung function tests.
In this study we presented a neuro-fuzzy system for classification of asthma and chronic obstructive pulmonary disease (COPD). According to GINA and GOLD guidelines we defined fuzzy rules and neural network parameters. ANN of system was trained on more than one thousand medical reports obtained from database of the company CareFusion. Implemented neuro-fuzzy system was validated on 455 patients by physicians from the Clinical Centre University of Sarajevo. All patients were separated into two groups, healthy and diseased. Diseased subjects were separated into two subgroups, asthmatics and COPD patients. Out of 170 asthmatic patients, neuro-fuzzy system correctly classified 99.41% of patients. In addition, out of 248 COPD patients 99.19% were correctly classified. The system was 100% correct on 37 patients with normal lung function. Based on our neuro-fuzzy system we obtained sensitivity of 99.28% and specificity of 100% in asthma and COPD classification. These results have been achieved due to the fact that in our neuro-fuzzy system are also implemented all recommendations of GINA and GOLD necessary for classification of asthma and COPD. Also, as shown in the results, in the process of establishing the final diagnosis, complete dynamic assessment of the patient is obtained, as opposed to the solution that provides a static assessment of the patient.
The authors would like to thanks to pulmological team from Clinical Centre University of Sarajevo, Bosnia and Herzegovina for their help during the research. We would also like to thank the company CareFusion that allowed us to access reports obtained from the measurements which we used for training NN and recommendations for better quality of software development.
Publication of this article was funded by authors.
This article has been published as part of BMC Medical Informatics and Decision Making Volume 15 Supplement 3, 2015: Multidimensional, multidiscipline and shared decision making in healthcare and eHealth. The full contents of the supplement are available online at http://www.biomedcentral.com/bmcmedinformdecismak/supplements/15/S3.
- Singh M: The burden of asthma in children: an Asian perspective. Pediatr Respir Rev. 2005, 6: 14-9. 10.1016/j.prrv.2004.11.003.View ArticleGoogle Scholar
- Sennhauser FH, Braun-Fahrlander C, Wildhaber JH: The burden of asthma in children: a European perspective. Pediatr Resp Rev. 2005, 6: 2-7. 10.1016/j.prrv.2004.11.001.View ArticleGoogle Scholar
- Beasley R: The burden of asthma with specific reference to the United States. J Allergy Clin Immunol. 2002, 5: 482-9.View ArticleGoogle Scholar
- Helms PJ: Issues and unmet needs in pediatric asthma. Pediatr Pulmonol. 2005, 30: 159-65.View ArticleGoogle Scholar
- Enright PL, McClelland RL, Newman AB, Gottlieb DJ, Lebowitz MD: Under diagnosis and under treatment of asthma in the elderly. Cardiovascular health study research group Chest. 1999, 116: 603-13.PubMedGoogle Scholar
- Tortora GJ, Grabowski SR: Principles of Anatomy and Physiology. 2003, New York, NY, USA, John Wiley & Sons, 10Google Scholar
- Mayo Clinic: COPD Definition. Accessed 14 Jan 2013, [http://www.mayoclinic.com/health/copd/DS00916]
- Anthonisen N: Chronic Obstructive Pulmonary Disease. Edited by: Goldman L, Auseillo D. Goldman. 2007, Cecil Medicine, Philadelphia, c:88-Google Scholar
- COPD Burden in Europe. Accessed 14 Jan 2013, [http://www.european-lung-foundation.org/index.php?id=63]
- Ferlay J: Cancer Incidence, Mortality and Prevalence Worldwide. IARC Cancer Base No.5, Version 2.0. 2004, IARC Press, Lyon, FranceGoogle Scholar
- Murray CJ, Lopez AD: Alternative projections of mortality and disability by cause 1990-2020. Global Burden of Disease Study. 1997, 349: 1498-504.Google Scholar
- Nielsen KG, Bisgaard H: The Effect of Inhaled Budesonide on Symptoms, Lung Function, and Cold Air and Methacholine Responsiveness in 2- to 5-year-old Asthmatic Children. Am J Respir Crit Care Med. 2000, 162: 1500-1506. 10.1164/ajrccm.162.4.2002019.View ArticlePubMedGoogle Scholar
- Song TW: Correlation between spirometry and impulse oscillometry in children with asthma. Acta Pediatr. 2008, 97: 51-54.View ArticleGoogle Scholar
- Barúa M, Nazeran H, Nava P, Granda V, Diong B: Classification of Pulmonary Diseases Based on Impulse Oscillometric Measurements of Lung Function Using Neural Networks. Proceedings of the 26th Annual International Conference of the IEEE EMBS. 2004, San Francisco, CA, USAGoogle Scholar
- Winkler J, Hagert-Winkler A, Wirtz H, Schauer J, Kahn T, Hoheisel G: Impulse oscillometry in the diagnosis of the severity of obstructive pulmonary disease. Pneumol. 2009, 63: 266-75. 10.1055/s-0028-1119698.View ArticleGoogle Scholar
- Asaithambi M, Manoharan SC, Subramanian S: Classification of Respiratory Abnormalities Using Adaptive Neuro Fuzzy Inference System. Int Inf and Database Sys Lecture Notes in Computer Science. 2012, 7198: 65-73. 10.1007/978-3-642-28493-9_8.View ArticleGoogle Scholar
- Meraz E, Nazeran H, Goldman M, Nava P, Diong B: Impulse Oscillometric Features of Lung Function: Towards Computer-Aided Classification of Respiratory Diseases in Children. 30th Annual International IEEE EMBS Conference Vancouver. 2008, British Columbia, CanadaGoogle Scholar
- Hafezi N: An integrated software package for model-based neuro-fuzzy classification of small airway dysfunction. ETD Collection for University of Texas, El Paso Paper AI1473869. 2009Google Scholar
- Global Initiative for Chronic Obstructive Lung Disease. Pocket guide to COPD Diagnosis, Management and Prevention. 2013Google Scholar
- Global Initiative for Asthma. Pocket guide to Asthma Diagnosis, Management and Prevention. Accessed 21 November 2014, [http://www.ginasthma.org/local/uploads/files/GINA_Report_2014_Aug12.pdf]
- Zolnoori M, Zarandi M, Moin M, Teimoran S: Fuzzy Rule-Based Expert System for Assessment Severity of Asthma. Journal of Medical Systems. 2010, 36 (3): 1707-1717.View ArticlePubMedGoogle Scholar
- Mishra N, Singh DBV, Bandil MK, Sharma P: Decision Support System for Asthma. International Journal of Information and Computation Technology. 2013, 3: 549-554.Google Scholar
- Uncu U, Koklukaya E, Gencsoy A: A fuzzy rule-base model for classification of spirometric FVC graphs in chronical obstructive pulmonary diseases. Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2001, 4: 3866-3869.Google Scholar
- Anand SK, Kalpana R, Vijayalakshmi S: Design and Implementation of a Fuzzy Expert System for Detecting and Estimating the Level of Asthma and Chronic Obstructive Pulmonary Disease. Middle-East Journal of Scientific Research. 2013, 14: 1435-1444.Google Scholar
- Fodor J: Left-continuous t-norms in fuzzy logic: An overview. Acta Polytech Hung. 2004, 1 (2):Google Scholar
- Khan J, Wei JS, Ringner M, Saal LH, Ladanyi M, Westermann F, Meltzer PS: Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nature medicine. 2001, 7 (6): 673-679. 10.1038/89044.View ArticlePubMedPubMed CentralGoogle Scholar
- Cevenini G, Severi FM, Bocchi C, Petraglia F, Barbini P: An informative probability model enhancing real time echobiometry to improve fetal weight estimation accuracy. Med Biol Eng Comput. 2008, 46 (2): 109-120. 10.1007/s11517-007-0299-2.View ArticlePubMedGoogle Scholar
- Güler NF, Übeyli ED, Güler İ: Recurrent neural networks employing Lyapunov exponents for EEG signals classification. Expert Systems with Applications. 2005, 29 (3): 506-514. 10.1016/j.eswa.2005.04.011.View ArticleGoogle Scholar
- Byvatov E, Fechner U, Sadowski J, Schneider G: Comparison of support vector machine and artificial neural network systems for drug/nondrug classification. Journal of Chem Inf and Comput Sci. 2003, 43 (6): 1882-1889. 10.1021/ci0341161.View ArticleGoogle Scholar
- Kermani BG, Schiffman SS, Nagle HT: Performance of the Levenberg-Marquardt neural network training method in electronic nose applications. Sensors and Actuators B: Chemical. 2005, 110 (1): 13-22. 10.1016/j.snb.2005.01.008.View ArticleGoogle Scholar
- Azemi A, Sabzevari VR, Khademi M, Gholizade H, Kiani A, Dastgheib ZS: Intelligent arrhythmia detection and classification using ICA. In Engineering in Medicine and Biology Society, 28th Annual International Conference of the IEEE. 2006, 2163-2166.Google Scholar
- Wang GCS, Jain CL: Regression Analysis Modelling & Forecasting. 2003, Graceway Publishing Company IncGoogle Scholar
- Hao Y, Wilamowski M: Levenberg-Marquardt Training. 2011, 1-16. eBook ISBN: 978-1-4398-0284-7, Print ISBN: 978-1-4398-0283-0Google Scholar
- Larsen GL: Impulse oscillometry versus spirometry in a long-term study of controller therapy for pediatric asthma. J Allergy Clin Immunol. 2009, 123: 861-867. 10.1016/j.jaci.2008.10.036.View ArticlePubMedGoogle Scholar
- Nikkhah M, Babak A, Eshaghian A, Fardad S, Asadian A, Roshanzamir T, Akbari M, Golshan M: Comparison of Impulse Oscillometry System and Spirometry for Diagnosis of Obstructive Lung Disorders. NRITLD, National Research Institute of Tuberculosis and Lung Disease, Iran, Tanaffos. 2011, 10 (1): 19-25.Google Scholar
- Kolsum U, Borrill Z, Roy K, Starkey C, Vestbo J, Houghton C: Impulse oscillometry in COPD: identification of measurements related to airway obstruction, airway conductance and lung volumes. Respir Med. 2009, 103 (1): 136-43. 10.1016/j.rmed.2008.07.014.View ArticlePubMedGoogle Scholar
- Veiga J, Lopes AJ, Jansen JM, Melo PL: Fluctuation analysis of respiratory impedance waveform in asthmatic patients: effect of airway obstruction. Med Biol Eng Comput. 2012, 50 (12): 1249-1259. 10.1007/s11517-012-0957-x.View ArticlePubMedGoogle Scholar
- Song TW, Kim KW, Kim ES, Kim KE, Sohn MH: Correlation between spirometry and impulse oscillometry in children with asthma. Acta Pediatr. 97 (1): 51-4. Erratum in: Acta Pediatr; 2011, 97 (4): 520Google Scholar
- Chakraborty C, Tamoghna M, Mukherjee A, Ray AK: CAIDSA: Computer-aided intelligent diagnostic system for bronchial asthma. Exp Sys App 36. 2009, 4958-4966.Google Scholar
- Price DB, Tinkelman DG, Halbert RJ, Nordyke RJ, Isonaka S, Nonikov D: Symptom-based questionnaire for identifying COPD in smokers. Respir. 2006, 73 (3): 285-294. 10.1159/000090142.View ArticleGoogle Scholar
- Tinkelman DG, Price DB, Nordyke RJ, Halbert RJ, Isonaka S, Nonikov D: Symptom-based questionnaire for differentiating COPD and asthma. Respir. 2006, 73 (3): 296-300. 10.1159/000090141.View ArticleGoogle Scholar
- Diong B: Modeling Human Respiratory Impedance: Comparing the Best Method with the Least Estimation Errors. IEEE Eng Med & Biol Magaz. 2009, 48-55.Google Scholar
- Diong B: The augmented RIC model of the human respiratory system. Med Biol Eng Comput, Vol 47. 2009, 4: 395-404.View ArticleGoogle Scholar
- Rajagiri A: Can Asthma in Children be Detected by the Estimated Parameter Values of the Augmented RIC Model?. Proceedings of the IEEE-Engineering in Medicine and Biology Society (EMBS), 28th Annual International Conference. 2006, 5595-5598.Google Scholar
- Rajagiri A: Evaluation of Augmented RIC Model of Adult Respiratory Impedance Based on Impulse Oscillometry Data. Proceedings of the 12th International Conference of Biomedical Engineering. 2005Google Scholar
- Rajagiri A, Diong B: Evaluation of Augmented RIC Model of Child Respiratory Impedance Based on Impulse Oscillometry Data. Proc IEEE Region 5 Technical Conference. 2006, 291-295.Google Scholar
- Baswa S: Evaluation of Respiratory System Models Based on Parameter Estimates from Impulse Oscillometry Data. Proceedings IEEE Engineering in Medicine and Biology 27th Annual Conference. 2005, 2958-2961.Google Scholar
- Nguyen T: A study of IOS Data Using Two Mead-related Models of Respiratory Impedance. Proceedings of the 29th Annual International Conference of the IEEE EMBS. 2007, 1078-1081.Google Scholar
- Nguyen T, Diong B: Evaluation of Two Mead-related Models of Adult Respiratory Impedance Based on IOS Data. IEEE Region 5 Technical Conference. 2007, 303-308.Google Scholar
- Badnjevic A, Cifrek M, Koruga D: Integrated software suite for diagnosis of respiratory diseases. Proceedings of the IEEE International Conference on Computer as Tool (EUROCON 2013), Zagreb, Croatia. 2013, 564-569.Google Scholar
- Badnjevic A, Koruga D, Cifrek M, Smith HJ, Bego T: Interpretation of pulmonary function test results in relation to asthma classification using integrated software suite. Proceedings of the 36th IEEE International convention on information and communication technology, electronics and microelectronics (MIPRO), Opatija, Croatia. 2013, 153-157.Google Scholar
- Badnjevic A, Cifrek M, Koruga D: Classification of Chronic Obstructive Pulmonary Disease (COPD) Using Integrated Software Suite. XIII Mediterranean Conference on Medical and Biological Engineering and Computing 2013, IFMBE Proceedings. 2014, 41: 911-914. 10.1007/978-3-319-00846-2_225.View ArticleGoogle Scholar
- Coppini G, Miniati M, Paterni M, Monti S, Ferdeghini EM: Computer-aided diagnosis of emphysema in COPD patients: Neural-network-based analysis of lung shape in digital chest radiographs. Medical Engineering & Physics. 2007, 29 (1): 76-86. 10.1016/j.medengphy.2006.02.001.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.