Usage of Weka Software Based On Machine Learning Algorithms for Prediction of Liver Fibrosis/Cirrhosis

Abstract

The liver, a life-sustaining organ, plays a substantial role in many body functions. Liver diseases have become an important world health problem in terms of prevalence, incidences, and mortalities. Liver fibrosis/cirrhosis is great of importance, because if not treated in time liver cancer could be occurred and spread to other parts of the body. For this reason, early diagnosis of liver fibrosis/cirrhosis gives significance. Accordingly, this study investigated the performances of different machine learning algorithms for prediction of liver fibrosis/cirrhosis based on demographic and blood values. In this context, random forest, k nearest neighbour, C4.5 decision tree, K-star, random tree and reduced error pruning tree algorithms were used. Two distinct approaches were employed to evaluate the performances of machine learning algorithms. In the first approach, the entire features of dataset were utilized, while in the second approach, only the features selected through principal component analysis were used. Each approach was rigorously assessed using both 10-fold cross-validation and data splitting (70% train and 30% test) techniques. By conducting separate evaluations for each approach, a comprehensive understanding of the effectiveness of utilizing all features versus extracted features based principal component analysis was attained, providing valuable insights into the impact of feature dimensionality reduction on model performance. In this study, all analyses were implemented on WEKA data mining tool. In the first approach, the classification accuracies of random forest algorithm were 89.72% and 90.75% with the application of data splitting (70%-30%) and cross-validation techniques, respectively. In the second approach, where feature reduction is performed using principal component analysis technique, the accuracy values obtained from data splitting and cross-validation techniques of random forest algorithm were 88.61% and 88.83%, respectively. The obtained results revealed out that random forest algorithm outperformed for both approaches. Besides, the application of principal component analysis technique negatively affected the classification performance of used machine learning algorithms. It is thought that the proposed model will guide specialist physicians in making appropriate treatment decisions for patients with liver fibrosis/cirrhosis, potentially leading to death in its advanced stages.

Keywords

• WEKA
• Liver fibrosis/cirrhosis
• Principal component analysis
• Feature selection
• Early diagnosis
• Machine learning

Usage of Weka Software Based on Machine Learning Algorithms for Prediction of Liver Fibrosis/Cirrhosis

Abstract

The liver, a life-sustaining organ, plays a substantial role in many body functions. Liver diseases have become an important world health problem in terms of prevalence, incidences, and mortalities. Liver fibrosis/cirrhosis is great of importance, because if not treated in time liver cancer could be occurred and spread to other parts of the body. For this reason, early diagnosis of liver fibrosis/cirrhosis gives significance. Accordingly, this study investigated the performances of different machine learning algorithms for prediction of liver fibrosis/cirrhosis based on demographic and blood values. In this context, random forest, k nearest neighbour, C4.5 decision tree, K-star, random tree and reduced error pruning tree algorithms were used. Two distinct approaches were employed to evaluate the performances of machine learning algorithms. In the first approach, the entire features of dataset were utilized, while in the second approach, only the features selected through principal component analysis were used. Each approach was rigorously assessed using both 10-fold cross-validation and data splitting (70% train and 30% test) techniques. By conducting separate evaluations for each approach, a comprehensive understanding of the effectiveness of utilizing all features versus extracted features based principal component analysis was attained, providing valuable insights into the impact of feature dimensionality reduction on model performance. In this study, all analyses were implemented on WEKA data mining tool. In the first approach, the classification accuracies of random forest algorithm were 89.72% and 90.75% with the application of data splitting (70%-30%) and cross-validation techniques, respectively. In the second approach, where feature reduction is performed using principal component analysis technique, the accuracy values obtained from data splitting and cross-validation techniques of random forest algorithm were 88.61% and 88.83%, respectively. The obtained results revealed out that random forest algorithm outperformed for both approaches. Besides, the application of principal component analysis technique negatively affected the classification performance of used machine learning algorithms. It is thought that the proposed model will guide specialist physicians in making appropriate treatment decisions for patients with liver fibrosis/cirrhosis, potentially leading to death in its advanced stages.

Keywords

• WEKA
• Liver fibrosis/cirrhosis
• Principal component analysis
• Feature selection
• Early diagnosis
• Machine learning

References

1. Acarlı K. 2020. Karaciğer sağlığını koruyan 10 hayati öneri. URL: https://www.memorial.com.tr/saglik-rehberi/karaciger-sagligini-koruyan-10-hayati-oneri (accessed date: August 28. 2023).
2. Alaybeyoğlu A, Mulayim N. 2018. Karaciğer kanseri teşhisinde destek vektör makinesi tabanli uzman sistem tasarimi. Tıp Teknolojileri Kongresi, 8-10 Kasım, Gazi Magaso, KKTC, ss: 208-210.
3. Alkuşak E, Gök M. 2014. Karaciğer yetmezliğinin teşhisinde makine öğrenmesi algoritmalarinin kullanimi. ISITES 2014, June 8-10, Karabük, Türkiye, pp: 703-707.
4. Asrani SK, Devarbhavi H, Eaton J, Kamath PS. 2019. Burden of liver diseases in the world. J Hepatol, 70(1): 151-171.
5. Azam MS, Rahman A, Iqbal SHS, Ahmed MT. 2020. Prediction of liver diseases by using few machine learning based approaches. Aust J Eng Innov Technol, 2(5): 85-90.
6. Breiman L. 2001. Random forests. Machine Learn, 45: 5-32.
7. Borulday MG, Yegin EG, Mahouti P, Gunes F. 2017. Diagnosing liver Diseases with decision tree algorithm. Inter J Tech Phys Problems Engin, 33: 67-70.
8. Bulut C, Ballı T, Yetkin EF. 2023. Filtre modelli öznitelik seçim algoritmalarının EEG tabanlı beyin bilgisayar arayüzü sistemindeki karşılaştırmalı sınıflandırma performansları. Gazi Üniv Müh Mimar Fak Derg 38(4): 2397-2408.

9. Del Campo JA, Gallego P, Grande L. 2018. Role of inflammatory response in liver diseases: Therapeutic strategies. World J Hepatol, 10(1): 1.
10. Dritsas E, Trigka M. 2023. Supervised machine learning models for liver disease risk prediction. Comput, 12(1): 19.
11. Duda RO, Hart PE, Stork DG. 2000. Pattern classification. Wiley, New Jersey, USA, 2nd ed., pp: 176-181.
12. Gaber A, Youness HA, Hamdy A, Abdelaal HM, Hassan AM. 2022. Automatic classification of fatty liver disease based on supervised learning and genetic algorithm. Applied Sci, 12(1): 521.
13. Gulia A, Vohra R, Rani P. 2014. Liver patient classification using intelligent techniques. Inter J Comput Sci Inform Technol 5(4): 5110-5115.
14. Işık K, Ulusoy SK. 2021. Metal Sektöründe üretim sürelerine etki eden faktörlerin veri madenciliği yöntemleriyle tespit edilmesi. Gazi Üniv Müh Mimar Fak Derg, 36(4): 1949-1962.
15. Kaya C, Erkaymaz O, Ayar O, Özer M. 2017 October. Classification of diabetic retinopathy disease from Video-Oculography (VOG): signals with feature selection based on C4. 5 decision tree. Medical Technologies National Congress (TIPTEKNO), 31 October-2 September, Trabzon, Türkiye, pp: 1-4.
16. Keleş A, Karslı ÖB, Keleş A. 2020. Makine öğrenme algoritmaları ile karaciğer hastalığının teşhisi. Turkish Stud Inform Technol Appl Sci, 15(1): 75-83.
17. Lin RH. 2009. An intelligent model for liver disease diagnosis. Artificial Intel Med, 47(1): 53-62.
18. Ma H, Xu CF, Shen Z, Yu CH, Li YM. 2018. Application of machine learning techniques for clinical predictive modeling: a cross-sectional study on nonalcoholic fatty liver disease in China. BioMed Res Inter, 2018: 4304376.
19. Mishra S, Tadesse Y, Dash A, Jena L, Ranjan P. 2021. Thyroid disorder analysis using random forest classifier. In Intelligent and Cloud Computing: Proceedings of ICICC 2019 Volume 2 Springer, Singapore, pp: 385-390.
20. Mukhyber SJ, Abdulah DA, Majeed AD. 2023. Classification of liver dataset using data mining algorithms. 1st International & 4th Local Conference for Pure Science (ICPS-2021), 26–27 May 2021, Diyala, Iraq, pp: 6.
21. Muthuselvan S, Rajapraksh S, Somasundaram K, Karthik K. 2018. Classification of liver patient dataset using machine learning algorithms. Int J Eng Technol, 7(3): 323.
22. Narin A, Kaya C, Pamuk Z. 2021. Automatic detection of coronavirus disease (covid-19): using x-ray images and deep convolutional neural networks. Pattern Anal Applicat, 24: 1207-1220.
23. Pahareeya J, Vohra R, Makhijani J, Patsariya S. 2014. Liver patient classification using intelligence techniques. Inter J Adv Res Comput Sci Software Engin, 4(2): 295-299.
24. Rahman AS, Shamrat FJM, Tasnim Z, Roy J, Hossain SA. 2019. A comparative study on liver disease prediction using supervised machine learning algorithms. Inter J Sci Technol Res, 8(11): 419-422.
25. Ramana BV, Babu MSP, Venkateswarlu NB. 2011. A critical study of selected classification algorithms for liver disease diagnosis. Inter J Database Manage Syst, 3(2): 101-114.
26. Schiff ER, Maddrey WC, Reddy KR. 2017. Schiff's Diseases of the Liver. John Wiley & Sons, London, UK, pp: 241.
27. Sevli O. 2019. Performance Comparison of Different Machine Learning Techniques in Diagnosis of Breast Cancer. Eur J Sci Technol, 16: 176-185.
28. Şenel FA, Saygin RR, Saygin M, Öztürk Ö. 2021. Makine Öğrenmesi Algoritmaları Kullanılarak Vücut Analizi ile Uyku Apnesi Teşhisi. Uyku Bülteni, 2(1): 6-10.
29. Ucar M, İncetaş MO. 2022. Classification of brain MRI using Efficientnet CNN model and feature selection method. İKSAD Publishing House, Ankara, Türkiye, pp: 110.
30. Uzun R, İşler Y, Toksan M. 2018 May. Choose of wart treatment method using Naive Bayes and k-nearest neighbors classifiers. 26th Signal Processing and Communications Applications Conference (SIU), 02-05 May 2018, İzmir, Türkiye, pp: 1-4.
31. Uzun R, İşler Y, Toksan M. 2019. WEKA yazılım paketinin siğil tedavi yöntemlerinin başarısının tahmininde kullanımı. Düzce Üniv Bilim Teknol Derg, 7(1): 699-708.
32. Ünal Y, Sağlam A, Kayhan O. 2019. Improving classification performance for an imbalanced educational dataset example using SMOTE. Avrupa Bilim Teknol Derg, 2019: 485-489.
33. Xie G, Wang X, Liu P, Wei R, Chen W, Rajani C, Jia W. 2016. Distinctly altered gut microbiota in the progression of liver disease. Oncotarget 7(15): 19355.
34. Velu SR, Ravi V, Tabianan K. 2022. Data mining in predicting liver patients using classification model. Health Technol, 12(6): 1211-1235.
35. Yapıcı Şenyer İ. 2021. Obezitenin elektroretinografi (ERG): sinyali üzerindeki etkisi. Doktora Tezi, Zonguldak Bülent Ecevit Üniversitesi, Fen Bilimleri Enstitüsü, Zonguldak,Türkiye, ss: 163.