Using the Artificial Intelligence Techniques for Diagnosing of intensity of Non-Alcoholic Fatty Liver Disease by Clinical Parameters
DOI:
https://doi.org/10.22100/jkh.v11i3.1369Keywords:
Disease diagnosis, Non-alcoholic fatty liver disease, Clinical parameters, Artificial neural network, Rule extraction.Abstract
Introduction: Non-Alcoholic Fatty Liver Disease (NAFLD) is one of the most prevalent liver diseases with several levels of severity. Recently, the FibroScan device has been used as a non-invasive method for measurement of elasticity of the liver and consequently its fattiness. The current study aimed at providing a low-cost and simple method for diagnosing of the disease through the clinical symptoms.
Methods: A collection of data obtained from 726 patients was used for conducting this study with each patient having a fatty liver disease with different levels of intensity. The severity of the disease for each patient was measured by FibroScan device along with medical tests and ultrasound monitoring. Then, the artificial neural networks were used for determination of the relationship between the data obtained from the patients and the intensity levels. Finally, by the aid of artificial intelligence techniques, a method is employed for extracting rules from artificial neural networks for representing the relationship between the data.
Results: According to the results obtained from FibroScan device, among the 726 patients, 5 were located in F4 class, 23 in F3 class, 132 in F2class, 151 in F1, and 415 in F0 Class (Healthy people). According to the proposed method, the accuracy of diagnosis for various classes is as follows: 100% for F4 class, 99.31% for F3 class, 93.94% for F2 class, and 80.58% for F1 Class. Accordingly, this method can identify the samples in F4 and F3 classes with an ideal accuracy and the samples in F1 and F2 classes with a good accuracy.
Conclusion: Results in this research indicate that the proposed method can be used for diagnosing NAFLD and identifying its intensity levels with a lower costs and easier accessibility. It extracts the required rules for diagnosing the dieses that can be used by the physicians in their diagnosis.
References
Jamali RJA. Non-alcoholic fatty liver disease. Journal of Kashan University of Medical Sciences 2010;14:169-81.[Persian].
Sug H, editors. Improving the prediction accuracy of liver disorder disease with oversampling. proceeding of the 6th wseas international conference on Computer Engineering and Application, and American conference on Applied Mathematics; 2012; Cambridge 2012.
Cales P, Boursier J, Chaigneau J, Laine F, Sandrini J, Michalak S, et al. Diagnosis of different liver fibrosis characteristics by blood tests in non-alcoholic fatty liver disease. Liver International 2010;30:1346-54. doi: 10.1111/j.1478-3231.2010.02314.x
Fujiwara Sh, Hongou Y, Miyaji K, Asai A, Tanabe T, Fukui H, et al. Relationship between liver fibrosis noninvasively measured by fibro scan and blood test. Bulletin of the Osaka Medical College 2007;53:93-105.
Janda JM ASL, editors. The Entrobacteria. 2nd ed. Washington: ASM press;2006.
Afdhal NH. Fibroscan (transient elastography) for the measurement of liver fibrosis. Gastroenterol Hepatol 2012;8:605-7.
Wai Ch, Greenson JK, Fontana RJ, Kalbfleisch JD, Marrero JA, Conjeevaram HS. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38:518-26. doi: 10.1053/jhep.2003.50346
Lok AS, Ghany MG, Goodman ZD, Wright EC, Everson GT, Sterling RK, et al. Predicting cirrhosis in patients with hepatitis c based on standard laboratory test: results of the halt-c cohort. Hepatology 2005;24:282-92. doi: 10.1002/hep.20772
Forns X, AmpurdanesS, Llovet JM, Aponte J, Quinto L, Martinez-Bauer E, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002;36:986-92. doi: 10.1053/jhep.2002.36128
Angulo P, Hui J, Marchesini G, Bugianesi E, George J, Farrel G, et al. The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology 2007;45:846-54. doi: 10.1002/hep.21496
Pournik O, Dorri S, Zabolinezhad H, Alavian SM, Eslami S. A diagnostic model for cirrhosis in patients with non-alcoholic fatty liver disease: an artificial neural network approach. Med J Islam Repub Iran 2014;28:116.
Roobeart D, Karakoulus G, editors. Information gain, correlation and support vector machine. In: Guyon I, Gunn S, editors. Feature extraction. Berlin: Springer-Verlag Berlin Heidelberg;2006.p.463-70. doi: 10.1007/978-3-540-35488-8
Cano A, Nguyen D, Ventura S, Cios K. ur-CAIM: improved CAIM discretization for unbalanced and balanced data. Soft Computing 2016;20:173-88. doi: 10.1007/s00500-014-1488-1
Antkowiak M. Artificial neural networks vs. Support vector machines for skin diseases recognition [dissertation]. Sweden: Umea Univ.;2006.
El Farissi I, Azizi M, Lanet J-L, Moussaoui M. Neural network vs. Bayesian network to detect Java card mutants. AASRI Procedia 2013;4:132-7. doi: 10.1016/j.aasri.2013.10.021
Ganesan N, Venkatesh K, Rama M, Palani AM. Application of neural networks in diagnosing cancer disease using demographic data. International Journal of Computer Applications 2010;1:76-85.
Jaleel JA, Salim S, Aswin R. Artificial neural network based skin cancer detection. International Journal of Advanced Research in Electrical 2012;1:200-5.
Khan IY, Zope P, Suralkar S. Importance of artificial neural network in medical diagnosis disease like acute nephritis disease and heart disease. International Journal of Engineering Science and Innovative Technology (IJESIT) 2013;2:210-7.
Cohen WW, editor. Fast effective rule induction. Proceedings of the 12th international conference on machine learning;1995; Tahoe City, California.
Quinlan JR. Learning with continuous classes. 5th Australian joint conference on artificial intelligence; 1992: Singapore.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
