Classification of diabetes disease using TCM electronic nose signals and ensemble learning

Diabetes is one of the most prevalent diseases in medical field. We propose an ensemble method for diagnosis of diabetes on traditional Chinese medicine electronic nose signals. To evaluate the effectiveness of our method, we carry out the experiments by comparing single classifier with ensemble classifiers based on support vector machine and logistic classification model. The proposed method shows better classification performance with accuracy of 88.04%. The results of this study show that ensemble method is effective to detect diabetes.

[1]  Babak Kateb,et al.  Sniffing out cancer using the JPL electronic nose: A pilot study of a novel approach to detection and differentiation of brain cancer , 2009, NeuroImage.

[2]  Naresh Magan,et al.  Electronic nose analysis of bronchoalveolar lavage fluid , 2011, European journal of clinical investigation.

[3]  Zsofia Lazar,et al.  Electronic Nose Breathprints Are Independent of Acute Changes in Airway Caliber in Asthma , 2010, Sensors.

[4]  T. Kerdcharoen,et al.  Diabetes diagnosis by direct measurement from urine odor using electronic nose , 2012, The 5th 2012 Biomedical Engineering International Conference.

[5]  Chengdong Wu,et al.  Classification of multi-channels SEMG signals using wavelet and neural networks on assistive robot , 2012, IEEE 10th International Conference on Industrial Informatics.

[6]  Zhang Li-wen,et al.  Appropriateness in Applying SVMs to Text Classification , 2010 .

[7]  Conrad Bessant,et al.  Electronic-Nose Technology Using Sputum Samples in Diagnosis of Patients with Tuberculosis , 2010, Journal of Clinical Microbiology.

[8]  Jian Liu,et al.  SVM algorithm based on wavelet kernel function for medical image segmentation , 2009, International Symposium on Multispectral Image Processing and Pattern Recognition.

[9]  S J Pöppl,et al.  Predicting Type 2 diabetes using an electronic nose-based artificial neural network analysis. , 2002, Diabetes, nutrition & metabolism.

[10]  E. Martinelli,et al.  Lung cancer identification by the analysis of breath by means of an array of non-selective gas sensors. , 2003, Biosensors & bioelectronics.

[11]  M. Phillips,et al.  Volatile Markers of Breast Cancer in the Breath , 2003, The breast journal.

[12]  Peter Elsner,et al.  Smelling Renal Dysfunction via Electronic Nose , 2005, Annals of Biomedical Engineering.

[13]  P. Wang,et al.  A novel method for diabetes diagnosis based on electronic nose. , 1997, Biosensors & bioelectronics.

[14]  Michael Phillips,et al.  Can the electronic nose really sniff out lung cancer? , 2005, American journal of respiratory and critical care medicine.

[15]  Zhezhou Zheng,et al.  Study on Application of Medical Diagnosis by Electronic Nose , 2012 .

[16]  Q. Zhang,et al.  Diagnosis of diabetes by image detection of breath using gas-sensitive LAPS. , 2000, Biosensors & bioelectronics.

[17]  Raed A. Dweik,et al.  Can the Electronic Nose Really Sniff out Lung Cancer , 2005 .

[18]  M. Ezzati,et al.  National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 1980: systematic analysis of health examination surveys and epidemiological studies with 370 country-years and 2·7 million participants , 2011, The Lancet.

[19]  Liu Yi-hui,et al.  Wavelet high-frequency coefficients for feature extraction of gene microarray data , 2011 .