Neural Computing and Applications

, Volume 31, Issue 4, pp 979–989 | Cite as

Rule extraction for fatty liver detection using neural networks

  • Mojtaba Shahabi
  • Hamid HassanpourEmail author
  • Hoda Mashayekhi
Original Article


Non-alcoholic fatty liver disease (NAFLD) is one of the most common diseases in the world. Recently the FibroScan device is used as a noninvasive, yet costly method to measure the liver’s elasticity as a NAFLD indicator. Other than the cost, the diagnosis is not widely accessible to all patients. On the other hand, early detection of the disease can prevent later risks. In this study, we aim to use learning methods to infer the NAFLD severity level, only based on clinical tests. A dataset was constructed from clinical and ultrasonography data of 726 patients who were diagnosed with different NAFLD severity levels. Artificial neural networks (ANN) were used to model the relationship between NAFLD and the clinical tests. Next, a method was used to analyze the ANN and extract compact and human understandable rules. The derived rules can detect the fatty liver disease with an accuracy above 80%.


Disease detection Non-alcoholic fatty liver disease (NAFLD) Artificial neural networks (ANNs) Rule extraction 


Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.


  1. 1.
    Stoean C, Stoean R, Lupsor M, Stefanescu H, Badea R (2011) Feature selection for a cooperative coevolutionary classifier in liver fibrosis diagnosis. Comput Biol Med 41(4):238–246CrossRefzbMATHGoogle Scholar
  2. 2.
    Acharya UR, Raghavendra U, Fujita H, Hagiwara Y, Koh JE, Hong TJ, Sudarshan VK, Vijayananthan A, Yeong CH, Gudigar A (2016) Automated characterization of fatty liver disease and cirrhosis using curvelet transform and entropy features extracted from ultrasound images. Comput Biol Med 79:250–258CrossRefGoogle Scholar
  3. 3.
    Sug H (2012) Improving the prediction accuracy of liver disorder disease with oversampling. In: Proceeding of the 6th WSEAS international conference on computer engineering and application, and American conference on applied mathematics, Cambridge, pp 331–335Google Scholar
  4. 4.
    Jiang ZG, Tapper EB, Connelly MA, Pimentel CM, Feldbrügge L, Kim M, Krawczyk S, Robson SC, Herman M, Otvos JD (2016) Steatohepatitis and liver fibrosis are predicted by the characteristics of very low density lipoprotein in nonalcoholic fatty liver disease. Liver Int 36(8):1213–1233CrossRefGoogle Scholar
  5. 5.
    Gorunescu F, Belciug S, Gorunescu M, Badea R (2012) Intelligent decision-making for liver fibrosis stadialization based on tandem feature selection and evolutionary-driven neural network. Expert Syst Appl 39(17):12824–12832CrossRefGoogle Scholar
  6. 6.
    Goceri E, Shah ZK, Layman R, Jiang X, Gurcan MN (2016) Quantification of liver fat: a comprehensive review. Comput Biol Med 71:174–189CrossRefGoogle Scholar
  7. 7.
    Siddiqui MS, Patidar KR, Boyett S, Luketic VA, Puri P, Sanyal AJ (2015) Performance of non-invasive models of fibrosis in predicting mild to moderate fibrosis in patients with nonalcoholic fatty liver disease (NAFLD). Liver Int 36:572–579CrossRefGoogle Scholar
  8. 8.
    Afdhal NH (2012) Fibroscan (transient elastography) for the measurement of liver fibrosis. Gastroenterol Hepatol 8(9):605Google Scholar
  9. 9.
    Gaia S, Campion D, Evangelista A, Spandre M, Cosso L, Brunello F, Ciccone G, Bugianesi E, Rizzetto M (2015) Non-invasive score system for fibrosis in chronic hepatitis: proposal for a model based on biochemical, FibroScan and ultrasound data. Liver Int 35(8):2027–2035CrossRefGoogle Scholar
  10. 10.
    Bril F, Ortiz-Lopez C, Lomonaco R, Orsak B, Freckleton M, Chintapalli K, Hardies J, Lai S, Solano F, Tio F (2015) Clinical value of liver ultrasound for the diagnosis of nonalcoholic fatty liver disease in overweight and obese patients. Liver Int 35(9):2139–2146CrossRefGoogle Scholar
  11. 11.
    Cales P, Boursier J, Chaigneau J, Laine F, Sandrini J, Michalak S (2010) Diagnosis of different liver fibrosis characteristics by blood tests in non-alcoholic fatty liver disease. Liver Int 30(9):1346–1354. doi: 10.1111/j.1478-3231.2010.02314.x CrossRefGoogle Scholar
  12. 12.
    Fujiwara Sh, Hongou Y, Miyaji K, Asai A, Tanabe T, Fukui H (2007) Relationship between liver fibrosis noninvasively measured by fibro scan and blood test. Bull Osaka Med Coll 35(2):93–105Google Scholar
  13. 13.
    Angulo P, Hui JM, Marchesini G, Bugianesi E, George J, Farrel GC (2007) The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology 45(4):846–854CrossRefGoogle Scholar
  14. 14.
    Forns X, Ampurdanes S, Llovet JM, Aponte J, Quinto L, Martinez-Bauer E (2002) Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 36(4):986–992CrossRefGoogle Scholar
  15. 15.
    Lok A, Ghany MG, Goodman ZD, Wright EC, Everson GT, Sterling RK (2005) Predicting cirrhosis in patients with hepatitis C based on standard laboratory test: results of the Halt-C cohort. Hepatology 24(2):282–292CrossRefGoogle Scholar
  16. 16.
    Wai Ch, Greenson JK, Fontana RJ, Kalbfleisch JD, Marrero JA, Conjeevaram HS (2003) A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 38(2):518–526CrossRefGoogle Scholar
  17. 17.
    Augasta MG, Kathirvalavakumar T (2012) Rule extraction from neural networks—a comparative study. In: Proceeding of the international conference on pattern recognition, informatics and medical engineering, Salem, Tamil Nadu. IEEE, pp 404–408. doi: 10.1109/ICPRIME.2012.6208380
  18. 18.
    Kahramanli H, Allahverdi N (2009) Extracting rules for classification problems: AIS based approach. Expert Syst Appl 36(7):10494–10502CrossRefGoogle Scholar
  19. 19.
    Kamruzzaman SM, Sarkar AM (2011) A new data mining schema using artificial neural networks. Sensors 11(5):4622–4647. doi: 10.3390/s110504622 CrossRefGoogle Scholar
  20. 20.
    Chorowski J, Zurada JM (2011) Extracting rules from neural networks as decision diagrams. IEEE Trans Neural Networks 22(12):2435–2446. doi: 10.1109/TNN.2011.2106163 CrossRefGoogle Scholar
  21. 21.
    Kulluk S, Özbakı L, Baykasoglu A (2013) Fuzzy DIFACONN-miner: a novel approach for fuzzy rule extraction from neural networks. Expert Syst Appl 40(3):938–946. doi: 10.1016/j.eswa.2012.05.050 CrossRefGoogle Scholar
  22. 22.
    Setiono R (1997) Extracting rules from neural network by pruning and hidden-unit node splitting. Neural Comput 9(1):205–225CrossRefzbMATHGoogle Scholar
  23. 23.
    Setiono R (2000) Extracting M-of-N rules from trained neural networks. IEEE Trans Neural Networks 11(2):512–519. doi: 10.1109/72.839020 CrossRefGoogle Scholar
  24. 24.
    Tsukimoto H (2000) Extracting rules form trained neural networks. IEEE Trans Neural Networks 11(2):377–389. doi: 10.1109/72.839008 CrossRefGoogle Scholar
  25. 25.
    Fu X, Wang L (2002) Rule extraction using a novel gradient-based method and data dimensionality reduction. In: Proceeding of the international joint conference on neural networks, Honolulu, HI, IEEE, pp 1275–1280. doi: 10.1109/IJCNN.2002.1007678
  26. 26.
    Kamruzzaman SM, Islam MD (2006) An algorithm to extract rules from artificial neural networks for medical diagnosis problems. Int J Inf Technol 12(8):41–59Google Scholar
  27. 27.
    Tewary G (2015) Effective data mining for proper mining classification using neural networks. Int J Data Min Knowl Manag Process 5(2):65–82CrossRefGoogle Scholar
  28. 28.
    Yu S, Guo X, Zhu K, Du J (2010) A neuro-fuzzy GA-BP method of seismic reservoir fuzzy rules extraction. Expert Syst Appl 37(3):2037–2042CrossRefGoogle Scholar
  29. 29.
    Wang J, Lim CP, Creighton D, Khorsavi A, Nahavandi S, Ugon J, Vamplew P, Stranieri A, Martin L, Freischmidt A (2015) Patient admission prediction using a pruned fuzzy min–max neural network with rule extraction. Neural Comput Appl 26(2):277–289CrossRefGoogle Scholar
  30. 30.
    Korosec M (2007) Technological information extraction of free form surfaces using neural networks. Neural Comput Appl 16(4–5):453–463CrossRefGoogle Scholar
  31. 31.
    Malone J, McGarry K, Wermter S, Bowerman Ch (2006) Data mining using rule extraction from kohonen self-organising maps. Neural Comput Appl 15(1):9–17CrossRefGoogle Scholar
  32. 32.
    Oh SK, Kim WD, Pedrycz W, Park BJ (2010) Polynomial-based radial basis function neural networks (P-RBF NNs) realized with the aid of particle swarm optimization. Fuzzy Sets Syst 163(1):54–77. doi: 10.1016/j.fss.2010.08.007 MathSciNetCrossRefzbMATHGoogle Scholar
  33. 33.
    Karabatak M, Ince MC (2009) An expert system for detection of breast cancer based on association rules and neural network. Expert Syst Appl 36(2):3465–3469CrossRefGoogle Scholar
  34. 34.
    Chan KY, Ling S-H, Dillon TS, Nguyen HT (2011) Diagnosis of hypoglycemic episodes using a neural network based rule discovery system. Expert Syst Appl 38(8):9799–9808CrossRefGoogle Scholar
  35. 35.
    Karthik S, Priyadarishini A, Anuradha J, Tripathy BK (2011) Classification and rule extraction using rough set for diagnosis of liver disease and its types. Adv Appl Sci Res 2(3):334–345Google Scholar
  36. 36.
    Fortuny EJ, Martens D (2012) Active learning based rule extraction for regression. Paper presented at the 12th international conference on data mining workshops (ICDMW), BrusselsGoogle Scholar
  37. 37.
    Widodo A, Shim M-C, Caesarendra W, Yang B-S (2011) Intelligent prognostics for battery health monitoring based on sample entropy. Expert Syst Appl 38(9):11763–11769CrossRefGoogle Scholar
  38. 38.
    Young WA II, Weckman GR (2010) Using a heuristic approach to derive a grey-box model through an artificial neural network knowledge extraction technique. Neural Comput Appl 19(3):353–366CrossRefGoogle Scholar
  39. 39.
    Kamruzzaman SM, Islam M (2007) Extraction of symbolic rules from artificial neural networks. Int J Comput Inf Sci Eng 1(10):3022–3028Google Scholar
  40. 40.
    Heh JS, Chen JC, Chang M (2008) Designing a decompositional rule extraction algorithm for neural networks with bound decomposition tree. Neural Comput Appl 17(3):297–309CrossRefGoogle Scholar
  41. 41.
    Plikynas DSL, Rasteniene A (2005) Portable rule extraction method for neural network decisions reasoning. Syst Cybern Inform 3(4):79–84Google Scholar
  42. 42.
    Siraj F, Omer EA, Hassan R (2012) Data mining and neural networks: the impact of data representation. In: A Karahoca (ed) Advances in data mining knowledge discovery and applications. InTech, Rijeka, pp 463–470. doi: 10.5772/51594 Google Scholar
  43. 43.
    Roobeart D, Karakoulus G, Chawla NV (2006) Information gain, correlation and support vector machine. In: Guyon I, Nikravesh M, Gunn S, Zadeh LA (eds) Feature extraction. Studies in fuzziness and soft computing, vol 207. Springer, Berlin, pp 463–470. doi: 10.1007/978-3-540-35488-8_23 Google Scholar
  44. 44.
    Kurgan LA, Cios KJ (2004) CAIM discretization algorithm. IEEE Trans Knowl Data Eng 16(2):145–153CrossRefGoogle Scholar
  45. 45.
    Kurgan LA, Cios, KJ (2003) Fast class-attribute interdependence maximization (CAIM) discretization algorithm. In: Proceedings of the 2003 international conference on machine learning and applications, Los Angeles, California, USA, 2003. CSREA Press, pp 30–36Google Scholar
  46. 46.
    Vora Sh, Mehta RG (2012) MCAIM: Modified CAIM discretization algorithm for classification. Int J Appl Inf Syst IJAIS 3(5):42–50Google Scholar
  47. 47.
    Cano A, Nguyen DT, Ventura S, Cios KJ (2014) ur-CAIM: Improved CAIM discretization for unbalanced and balanced data. Soft Comput. doi: 10.1007/s00500-014-1488-1 Google Scholar
  48. 48.
    Cohen WW (1995) Fast effective rule induction. In: Proceedings of the twelfth international conference on machine learning (1995), pp 115–123 Key: citeulike:3157878, Tahoe City, California, USA, 1995. pp 115–123Google Scholar
  49. 49.
    Quinlan JR (1992) Learning with continuous classes. In: 5th Australian joint conference on artificial intelligence, Singapore, pp 343–348Google Scholar
  50. 50.
    Lichman M (2013) UCI machine learning repository. Irvine, CA, University of California, School of Information and Computer Science.
  51. 51.
    Farquad MAH, Sultana J, Nagalaxmi G, Savankumar G (2014) Knowledge discovery from data: comparative study. Trans Eng Sci 2(6):72–75Google Scholar
  52. 52.
    Kamruzzaman SM (2007) RGANN: An efficient algorithm to extract rules from ANNs. J Electron Comput Sci 8:19–30Google Scholar
  53. 53.
    Kaczmar UM, Trelak W (2005) Fuzzy logic and evolutionary algorithm—two techniques in rule extraction from neural networks. Neurocomputing 63:359–379CrossRefGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2017

Authors and Affiliations

  • Mojtaba Shahabi
    • 1
  • Hamid Hassanpour
    • 1
    Email author
  • Hoda Mashayekhi
    • 1
  1. 1.Faculty of Computer Engineering and ITShahrood University of TechnologyShahroodIran

Personalised recommendations