Feature extraction-based liver tumor classification using Machine Learning and Deep Learning methods of computed tomography images

Figures & data

Figure 1. Tumorous and non-tumorous liver CT images dataset.

Figure 2. Histogram Equalization (a) Liver image before HE. (b) Pre HE graph, (c) Liver image after HE, (d) Post HE graph (Ciecholewski, 2007).

Figure 3. EMLTI liver tumor classification framework.

Table 1. Accuracy of ML and DL classifiers using all features.

ML classifiers	Accuracy (%)	DL classifiers	Accuracy (%)
DT	96.5	NN	50.0
RF	99.6	LSTM	53.0
Boost	99.7	Bi-LSTM	54.0
SVM	98.0	CNN	54.0

Figure 4. Accuracy chart of ML and DL with complete feature list.

Table 2. Accuracy of ML and DL classifiers using co-occurrence.

ML classifiers	Accuracy (%)	DL classifiers	Accuracy (%)
DT	96.5	NN	93.0
RF	99.6	LSTM	0.0
Boost	99.7	Bi-LSTM	49.0
SVM	98.0	CNN	49.0

Figure 5. Accuracy chart of ML and DL with co-occurrence feature list.

Table 3. Accuracy of ML and DL classifiers using wavelet transform features.

ML classifiers	Accuracy (%)	DL classifiers	Accuracy (%)
DT	96.5	NN	77.0
RF	99.6	LSTM	47.0
Boost	99.7	Bi-LSTM	49.0
SVM	98.0	CNN	54.0

Figure 6. Accuracy chart of ML and DL with wavelet transform feature list.

Table 4. Accuracy of ML and DL classifiers using gradient features.

ML classifiers	Accuracy (%)	DL classifiers	Accuracy (%)
DT	96.5	NN	82.0
RF	99.6	LSTM	49.0
Boost	99.7	Bi-LSTM	49.0
SVM	98.0	CNN	47.0

Figure 7. Accuracy chart of ML and DL with gradient feature list.

Table 5. Accuracy of ML and DL classifiers using autogressive features.

ML classifiers	Accuracy (%)	DL classifiers	Accuracy (%)
DT	96.5	NN	97.0
RF	99.6	LSTM	48.0
Boost	99.7	Bi-LSTM	48.0
SVM	98.0	CNN	49.0

Figure 8. Accuracy chart of ML and DL with autogressive feature list.

Table 6. Accuracy chart of ML and DL with complete and separate feature list.

Classifiers	Accuracy (%)	Accuracy (%)	Accuracy (%)	Accuracy (%)	Accuracy (%)
	Complete features	Co-occurrence features	Co-wavelet transform features	Gradient features	Autogressive features
DT	96.5	95.0	30.0	75.0	92.0
RF	99.6	99.0	98.0	99.6	98.0
Boost	99.7	99.0	98.0	87.0	98.0
SVM	98.0	99.0	96.0	86.0	96.0
NN	50.0	93.0	77.0	82.0	97.0
LSTM	53.0	0.0	47.0	49.0	48.0
Bi-LSTM	54.0	49.0	49.0	49.0	48.0
CNN	54.0	49.0	54.0	47.0	49.0

Figure 9. Frequency table achieved by ML and DL on different feature sets.

Table 7. Accuracy of pre-trained DL models.

Pre-trained DL classifiers	Accuracy (%)
VGG16	88.0
LSTM + CNN	97.0
Resnet50	78.0

Figure 10. Accuracy chart of pre-trained DL models.

Table 8. State-of-the-art advancement in liver tumor classification.

Reference	Description	Accuracy
(Kumar et al. 2010)	Fuzzy C mean Clustering, Adaptive Thresholding	94.3%
(Wu et al. 2016)	Volume of Interest (VOI), Linear Iterative Clustering Approach	96.00%
(Alahmer et al. 2016)	Fuzzy C mean Clustering, SVM, Multiple ROIs	91.63%
Chang et al. (2017)	Binary Logistic Regression Analysis. Leave one out Cross Validation	81.00%
(Choi et al. 2018)	vAlexNet, VGG16,	80.00%
(Anter & Hassenian 2019)	Watershed Algorithm, Adaptive Threshold, Morphological Operationsv	95.08%
(Parsai et al. 2019)	HERMES, Likert Score, Histopathology	94.7%
(Xu et al. 2019)	Binary Logistic regression, Texture Parameters, GLCM	89.0%
(Krishan & Mittal, 2021)	Liver cancer classification, k-fold cross validation, multilevel ensemble model	87.01%
MLTI Framework (Hussain et al., 2022)	Multiclass Features Dataset & ML Algorithms	97.48%

Ciecholewski, M. a. (2007) Automatic segmentation of single and multiple neoplastic hepatic lesions in CT images [Paper presentation]. In International Work-Conference on the Interplay between Natural and Artificial Computation, 63–71.

 Google Scholar

Kumar, S. S., & Moni, R. S. (2010). Diagnosis of liver tumor from CT images using curvelet transform. International Journal on Computer Science and Engineering, 2(4), 1173–1178.

 Google Scholar

Wu, W., Zhou, Z., Wu, S., & Zhang, Y. (2016). Automatic liver segmentation on volumetric CT images using supervoxel-based graph cuts. Computational and Mathematical Methods in Medicine, 2016, 9093721–9093714. 2016. https://doi.org/10.1155/2016/9093721

 PubMed Web of Science ®Google Scholar

Alahmer, H., & Ahmed, A. (2016). Hierarchical classification of liver tumor from CT images based on difference-of-features (DOF). World Congress on Engineering.

 Google Scholar

Chang, C.-C., Chen, H.-H., Chang, Y.-C., Yang, M.-Y., Lo, C.-M., Ko, W.-C., Lee, Y.-F., Liu, K.-L., & Chang, R.-F. (2017). Computer-aided diagnosis of liver tumors on computed tomography images. Computer Methods and Programs in Biomedicine, 145, 45–51. https://doi.org/10.1016/j.cmpb.2017.04.008

 PubMed Web of Science ®Google Scholar

Choi, K. J., Jang, J. K., Lee, S. S., Sung, Y. S., Shim, W. H., Kim, H. S., Yun, J., Choi, J.-Y., Lee, Y., Kang, B.-K., Kim, J. H., Kim, S. Y., & Yu, E. S. (2018). Development and validation of a deep learning system for staging liver fibrosis by using contrast agent–enhanced CT images in the liver. Radiology, 289(3), 688–697. https://doi.org/10.1148/radiol.2018180763

 PubMed Web of Science ®Google Scholar

Anter, A. M., & Hassenian, A. E. (2019). CT liver tumor segmentation hybrid approach using neutrosophic sets, fast fuzzy c-means and adaptive watershed algorithm. Artificial Intelligence in Medicine, 97, 105–117. https://doi.org/10.1016/j.artmed.2018.11.007

 PubMed Web of Science ®Google Scholar

Parsai, A., Miquel, M. E., Jan, H., Kastler, A., Szyszko, T., & Zerizer, I. (2019). Improving liver lesion characterisation using retrospective fusion of FDG PET/CT and MRI. Clinical Imaging, 55, 23–28. https://doi.org/10.1016/j.clinimag.2019.01.018

 PubMed Web of Science ®Google Scholar

Xu, X.-F., Xing, H., Han, J., Li, Z.-L., Lau, W.-Y., Zhou, Y.-H., Gu, W.-M., Wang, H., Chen, T.-H., Zeng, Y.-Y., Li, C., Wu, M.-C., Shen, F., & Yang, T. (2019). Risk factors, patterns, and outcomes of late recurrence after liver resection for hepatocellular carcinoma: A multicenter study from China. JAMA Surgery, 154(3), 209–217. https://doi.org/10.1001/jamasurg.2018.4334

 PubMed Web of Science ®Google Scholar

Krishan, A., & Mittal, D. (2021). Ensembled liver cancer detection and classification using CT images. Proceedings of the Institution of Mechanical Engineers. Part H, Journal of Engineering in Medicine, 235(2), 232–244. https://doi.org/10.1177/0954411920971888

 PubMed Web of Science ®Google Scholar

Hussain, M., Saher, N., & Qadri, S. (2022). Computer vision approach for liver tumor classification using CT dataset. Applied Artificial Intelligence, 36(1), 2055395. https://doi.org/10.1080/08839514.2022.2055395

 Web of Science ®Google Scholar

Code availability

The code will be available upon request to the corresponding author.

Data availability

The data will be available upon request to the corresponding author.