An intelligent prognostic system for analyzing patients with paraquat poisoning using arterial blood gas indexes.

The arterial blood gas (ABG) test is used to assess gas exchange in the lung, and the acid-base level in the blood. However, it is still unclear whether or not ABG test indexes correlate with paraquat (PQ) poisoning. This study investigates the predictive value of ABG tests in prognosing patients with PQ poisoning; it also identifies the most significant indexes of the ABG test. An intelligent machine learning-based system was established to effectively give prognostic analysis of patients with PQ poisoning based on ABG indexes. In the proposed system, an enhanced support vector machine combined with a feature selection strategy was developed to predict the risk status from a pool of 103 patients (56 males and 47 females); of these, 52 subjects were deceased and 51 patients were alive. The proposed method was rigorously evaluated against the real-life dataset in terms of accuracy, sensitivity, and specificity. Additionally, the feature selection was investigated to identify correlating factors for the risk status. The results demonstrated that there were significant differences in ABG indexes between deceased and alive subjects (p-value <0.01). According to the feature selection, we found that the most important correlated indexes were associated with partial pressure of carbon dioxide (PCO2). This study discovered the relationship between ABG test and poisoning degree to provide a new avenue for prognosing PQ poisoning.

[1]  Kyung-shik Shin,et al.  An application of support vector machines in bankruptcy prediction model , 2005, Expert Syst. Appl..

[2]  Gang Wang,et al.  Towards an optimal support vector machine classifier using a parallel particle swarm optimization strategy , 2014, Appl. Math. Comput..

[3]  Ilias Maglogiannis,et al.  An intelligent system for automated breast cancer diagnosis and prognosis using SVM based classifiers , 2009, Applied Intelligence.

[4]  Elif Derya Übeyli.  Comparison of different classification algorithms in clinical decision‐making , 2007, Expert Syst. J. Knowl. Eng..

[5]  Gang Wang,et al.  A new hybrid method based on local fisher discriminant analysis and support vector machines for hepatitis disease diagnosis , 2011, Expert Syst. Appl..

[6]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[7]  Sung-Chul Yoon,et al.  Clinical Outcome of Paraquat Poisoning , 2009, The Korean journal of internal medicine.

[8]  Gang Wang,et al.  Support Vector Machine Based Diagnostic System for Breast Cancer Using Swarm Intelligence , 2012, Journal of Medical Systems.

[9]  M. Cho,et al.  The prognostic value of arterial blood gas analysis in high‐risk acute heart failure patients: an analysis of the Korean Heart Failure (KorHF) registry , 2015, European journal of heart failure.

[10]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[11]  Xianqin Wang,et al.  Clearance Rate and BP-ANN Model in Paraquat Poisoned Patients Treated with Hemoperfusion , 2015, BioMed research international.

[12]  Changbao Huang,et al.  Prognostic significance of arterial blood gas analysis in the early evaluation of Paraquat poisoning patients , 2011, Clinical toxicology.

[13]  Seyed Mohammad Seyed Saadat,et al.  Prognostic value of arterial blood gas disturbances for in-hospital mortality in pediatric patients with severe traumatic brain injury , 2013, Acta Neurochirurgica.

[14]  Dayou Liu,et al.  A support vector machine classifier with rough set-based feature selection for breast cancer diagnosis , 2011, Expert Syst. Appl..

[15]  M. Craske,et al.  Ambulatory monitoring of panic patients during regular activity: A preliminary report , 1995, Biological Psychiatry.

[16]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[17]  Yu Cao,et al.  APACHE Score, Severity Index of Paraquat Poisoning, and Serum Lactic Acid Concentration in the Prognosis of Paraquat Poisoning of Chinese Patients , 2015, Pediatric emergency care.

[18]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[19]  Gang Wang,et al.  A Three-Stage Expert System Based on Support Vector Machines for Thyroid Disease Diagnosis , 2012, Journal of Medical Systems.

[20]  E. Pishbin,et al.  The correlation between end-tidal carbon dioxide and arterial blood gas parameters in patients evaluated for metabolic acid-base disorders , 2015, Electronic physician.

[21]  A. Soloukides,et al.  A Fatal Case of Paraquat Poisoning Following Minimal Dermal Exposure , 2007, Renal failure.

[22]  Steven Salzberg,et al.  On Comparing Classifiers: Pitfalls to Avoid and a Recommended Approach , 1997, Data Mining and Knowledge Discovery.

[23]  Chih-Jen Lin,et al.  Asymptotic Behaviors of Support Vector Machines with Gaussian Kernel , 2003, Neural Computation.

[24]  David G. Stork,et al.  Pattern Classification , 1973 .

[25]  Mehmet Fatih Akay,et al.  Support vector machines combined with feature selection for breast cancer diagnosis , 2009, Expert Syst. Appl..

[26]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[27]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.