ENHANCING LIVER DISEASE DIAGNOSIS THROUGH TEXTURE-BASED CLASSIFICATION OF CT IMAGES: A COMPUTER-AIDED DIAGNOSTIC APPROACH

Main Article Content

Fatima Hussain
Mehrun Nisa
Marina Riasat
Wajeha Shafqat
Nimra Batool
Faheem Afzal
Aalia Nazir
Ambreen Kalsoom

Keywords

Texture analysis, CT images, MaZda, PCA, LDA, NDA, POE+ACC

Abstract

Background: The incident of liver diseases among patients has been consistently increasing. Early detection and precise differentiation between benign and metastasis liver tumours could potentially lead to enhance the success rate. Computed tomography imaging serves as a practical medical imaging method for evaluating liver tumours.


Objective: The primary aim of this study is to classify liver CT images using texture features, with the goal of assessing the effectiveness of a computer-aided diagnostic system for detection of liver diseases.


Methods: The dataset used in this study comprises 162 cases of benign liver conditions and 128 cases of metastatic liver conditions. Statistical texture analysis techniques, such as Gray level run length matrix, and Co-occurrence matrix are employed to extract the texture features parameters from each region of interest. A normalization approach of μ ± 3σ is applied during this process to analyze the results, the Artificial Neural Network (ANN) classification method is utilized on Principal Component Analysis (PCA), Linear Discrimination Analysis (LDA) and Non- Linear Discrimination Analysis (NDA) techniques.


Results: The images are subjected to analysis using the analysis option available in MaZda software. By applying statistical features such as run length and gray level histogram, with normalization μ ± 3σ, the best results are obtained. This approach highlights the distinct difference between benign and metastasis liver diseases

Abstract 202 | pdf Downloads 57

References

1. Krishna, M., Microscopic anatomy of the liver. Clinical liver disease, 2013. 2(Suppl 1): p. S4.
2. Kermani, M.Z. and A. Gharbali, Automated differentiation of benign and malignant liver tumors by Ultrasound Images. The Journal of Urmia University of Medical Sciences, 2018. 29: p. 7.
3. Abdel-Misih, S.R. and M. Bloomston, Liver anatomy. Surgical Clinics, 2010. 90(4): p. 643-653.
4. López-Luque, J. and I. Fabregat, Revisiting the liver: from development to regeneration-what we ought to know! International Journal of Developmental Biology, 2018. 62(6-7-8): p. 441-451.
5. Qian, H., et al., Autophagy in liver diseases: A review. Molecular aspects of medicine, 2021. 82: p. 100973.
6. Ramana, B.V., M.S.P. Babu, and N. Venkateswarlu, A critical study of selected classification algorithms for liver disease diagnosis. International Journal of Database Management Systems, 2011. 3(2): p. 101-114.
7. Cheng, K.-C., et al., Association of different types of liver disease with demographic and clinical factors. Biomedicine, 2016. 6(3): p. 1-7.
8. Gunasundari, S. and S. Janakiraman, A study of textural analysis methods for the diagnosis of liver diseases from abdominal computed tomography. International Journal of Computer Applications, 2013. 74(11).
9. Gharbali, A., The Application of Texture Analysis for the Study of Liver Disease by Magnetic Resonance Imaging. 2007, University of Dundee.
10. Yue, Y., et al., Three-dimensional CT texture analysis to differentiate colorectal signet-ring cell carcinoma and adenocarcinoma. Cancer Management and Research, 2019. 11: p. 10445.
11. Summers, R.M., Texture analysis in radiology: does the emperor have no clothes? Abdominal Radiology, 2017. 42(2): p. 342-345.
12. Buch, K., et al., Using texture analysis to determine human papillomavirus status of oropharyngeal squamous cell carcinomas on CT. American Journal of Neuroradiology, 2015. 36(7): p. 1343-1348.
13. Huang, Z., et al., Two-dimensional texture analysis based on CT images to differentiate pancreatic lymphoma and pancreatic adenocarcinoma: a preliminary study. Academic radiology, 2019. 26(8): p. e189-e195.
14. Saleh, A. and R. Lerski, Texture Discrimination Analysis between Tumour and Healthy Brain Tissue Using MaZda Program. Al-Nahrain Journal of Science, 2011. 14(4): p. 57-65.
15. Ţălu, Ş., Texture analysis methods for the characterisation of biological and medical images. Extreme Life, Biospeology and Astrobiology, 2012. 4(1): p. 8-12.
16. Nisa, M., et al., Machine vision-based Statistical texture analysis techniques for characterization of liver tissues using CT images. 2022. 72(9): p. 1760-1765.
17. Li, Y., et al., CT image-based texture analysis to predict microvascular invasion in primary hepatocellular carcinoma. Journal of Digital Imaging, 2020. 33: p. 1365-1375.
18. Strzelecki, M., P. Szczypinski, A. Materka, and A. Klepaczko, A software tool for automatic classification and segmentation of 2D/3D medical images. Nuclear instruments and methods in physics research section A: Accelerators, Spectrometers, Detectors and associated equipment, 2013. 702: p. 137-140.
19. Gong, X., et al., Noninvasive assessment of significant liver fibrosis in rabbits by spectral CT parameters and texture analysis. Japanese Journal of Radiology, 2023: p. 1-11.
20. Nisa, M., S.A. Buzdar, K. Khan, and M.S. Ahmad, Deep convolutional neural network based analysis of liver tissues using computed tomography images. Symmetry, 2022. 14(2): p. 383.
21. Medrano, L.M., M.A. Jiménez-Sousa, A. Fernández-Rodríguez, and S. Resino, Genetic Polymorphisms Associated with Liver Disease Progression in HIV/HCV-Coinfected Patients. AIDS reviews, 2017. 19(1): p. 3-15.
22. Hamid, K., A. Asif, W. Abbasi, and D. Sabih. Machine learning with abstention for automated liver disease diagnosis. in 2017 International Conference on Frontiers of Information Technology (FIT). 2017. IEEE.
23. Nisa, M., et al., Machine vision-based Statistical texture analysis techniques for characterization of liver tissues using CT images. JPMA. The Journal of the Pakistan Medical Association, 2022. 72(9): p. 1760-1765.
24. Kumar, P., R. Kumar, and M.V. Reddy, Assessment of sewage treatment plant effluent and its impact on the surface water and sediment quality of river Ganga at Kanpur. Int J Sci Eng Res, 2017. 8(1): p. 1315-1324.
25. Mala, K., V. Sadasivam, and S. Alagappan, Neural network based texture analysis of CT images for fatty and cirrhosis liver classification. Applied Soft Computing, 2015. 32: p. 80-86.
26. Stocker, D., et al., MRI texture analysis for differentiation of malignant and benign hepatocellular tumors in the non-cirrhotic liver. Heliyon, 2018. 4(11): p. e00987.
27. Waite, S., et al., Communication errors in radiology–Pitfalls and how to avoid them. Clinical Imaging, 2018. 51: p. 266-272.