Machine Learning and Vision Transformer for CT Scanners' Calibration and Quality Assessment

Abstract

In this study, we present the process and research for finding the best machine learning methodology and innovative approach to evaluate the image quality in Computed Tomography (CT) scanners by predicting Signal-to-Noise Ratio (SNR) and Contrast-to-Noise Ratio (CNR) from low-resolution CT images of a series of phantoms. Traditional methods of Image Quality Assessment (IQA), reliant on subjective evaluation by radiologists, often suffer from variability and inefficiency. To address these limitations, we explored both interpretable models like the Adaptive Neuro-Fuzzy Inference System (ANFIS) and other advanced deep learning architectures. Initially, ANFIS combined with Gray Level Co-occurrence Matrix (GLCM) features yielded suboptimal results, with an R-squared value of 0.634. Experimenting with various deep learning methodologies for improving the performance, directed us to develop a hybrid model integrating DenseNet, Vision Transformers, and reparameterization techniques, which showed that can achieve superior results with an R-squared value of 0.8892. This research paper focuses on searching for the optimal machine learning model and lays the groundwork for an automated tool that can optimize imaging protocols by providing a comprehensive quality assessment of CT images in CT calibration.

Keywords

• Machine learning
• Deep learning
• Vision Transformer
• CT calibration

References

1. [1] Soufian, M., Robinson F.V.P., and Soufian M.,1996, Fuzzy Logic Controller for Whole Body NMR imaging. IEE Colloquium on Fuzzy Logic Controllers in Practice (Digest No. 96/200). London, U.K, DOI: 10.1049/ic:19961125.
2. [2] Bright J. J., Claydon M. A., Soufian M., and Gordon D. B., 2002, Rapid typing of bacteria using Matrix-Assisted Laser Desorption Ionisation Time-of-Flight Mass Spectrometry and Pattern Recognition Software, Journal of Microbiological Methods, Vol. 48, Issue 2-3, pp 127-138, https://doi.org/10.1016/S0167-7012(01)00317-7 . PMID:11777563
3. [3] Kumar, R. and Rattan, M., 2012. Analysis of various quality metrics for medical image processing. International Journal of Advanced Research in Computer Science and Software Engineering, 2(11), pp.137-144.
4. [4] Wang, C.L., Wang, C.M., Chan, Y.K. and Chen, R.T., 2012. Image‐quality figure evaluator based on contrast‐detail phantom in radiography. The International Journal of Medical Robotics and Computer Assisted Surgery, 8(2), pp.169-177.
5. [5] Khanh, Q. M., Alsaeedi, A. M., Soufian, M., Fulford, J. and Razak, A. H., 2024. SynQ-ViT: Synthetic Image Quality Assessment for CT Calibration with Vision Transformer, Submitted to IEEE-Engineering in Medicine & Biology Society Conference on Biomedical Engineering and Science (IECBES2024), Penang, Malaysia.
6. [6] Valdes, G., Scheuermann, R., Hung, C.Y., Olszanski, A., Bellerive, M. and Solberg, T.D., 2016. A mathematical framework for virtual IMRT QA using machine learning. Medical physics, 43(7), pp.4323-4334.
7. [7] Sharma, M. and Mukharjee, S., 2012. Artificial neural network fuzzy inference system (ANFIS) for brain tumor detection. arXiv preprint arXiv:1212.0059, pp.1-5.R. E. Sorace, V. S. Reinhardt, and S. A. Vaughn, “High-speed digital-to-RF converter,” U.S. Patent 5 668 842, Sept. 16, 1997.
8. [8] Hossam, A., Fawzy, A., Elnaghi, B.E. and Magdy, A., 2022. An intelligent model for rapid diagnosis of patients with COVID-19 based on ANFIS. In Proceedings of the International Conference on Advanced Intelligent Systems and Informatics 2021 (pp. 338-355). Springer International Publishing.

9. [9] Bahonar, B.M., Changizi, V., Ebrahiminia, A. and Baradaran, S., 2023. Prediction of breast dose in chest CT examinations using adaptive neuro-fuzzy inference system (ANFIS). Physical and Engineering Sciences in Medicine, 46(3), pp.1071-1080.
10. [10] Lee, J.H., Grant, B.R., Chung, J.H., Reiser, I. and Giger, M., 2018, March. Assessment of diagnostic image quality of computed tomography (CT) images of the lung using deep learning. In Medical Imaging 2018: Physics of Medical Imaging (Vol. 10573, pp. 399-405). SPIE.
11. [11] Shi, Y., Xia, W., Wang, G. and Mou, X., 2024. Blind ct image quality assessment using ddpm-derived content and transformer-based evaluator. IEEE Transactions on Medical Imaging.
12. [12] Jensen, C.T., Liu, X., Tamm, E.P., Chandler, A.G., Sun, J., Morani, A.C., Javadi, S. and Wagner-Bartak, N.A., 2020. Image quality assessment of abdominal CT by use of new deep learning image reconstruction: initial experience. American Journal of Roentgenology, 215(1), pp.50-57.
13. [13] Jang, J.S., 1993. ANFIS: adaptive-network-based fuzzy inference system. IEEE transactions on systems, man, and cybernetics, 23(3), pp.665-685.
14. [14] Korchiyne, R., Farssi, S.M., Sbihi, A., Touahni, R. and Alaoui, M.T., 2014. A combined method of fractal and GLCM features for MRI and CT scan images classification. arXiv preprint arXiv:1409.4559.
15. [15] Ramamurthy, B. and Chandran, K.R., 2012. Content based medical image retrieval with texture content using gray level co-occurrence matrix and k-means clustering algorithms. Journal of Computer Science, 8(7), p.1070.
16. [16] Soufian M., Molaei M., and Nefti S., 2017, Adaptive clustering based inclusion and computational intelligence for fed-batch fermentation process control. In IEEE Development in eSystem Engineering (DeSE), Paris, France. DOI: 10.1109/DeSE.2017.45.
17. [17] Targ, S., Almeida, D. and Lyman, K., 2016. Resnet in resnet: Generalizing residual architectures. arXiv preprint arXiv:1603.08029.
18. [18] Xu, W., Fu, Y.L. and Zhu, D., 2023. ResNet and its application to medical image processing: Research progress and challenges. Computer Methods and Programs in Biomedicine, 240, p.107660.
19. [19] Grossberg, S., 2013. Recurrent neural networks. Scholarpedia, 8(2), p.1888.
20. [20] Zhang, H. and Qie, Y., 2023. Applying deep learning to medical imaging: a review. Applied Sciences, 13(18), p.10521.
21. [21] Thiruppathi, K., Selvakumar, K. and Shenbagavel, V., 2023. SERESNET: Monkeypox Detection Model. International Journal of Advanced Computer Science and Applications, 14(9).
22. [22] Abdelrahman, A. and Viriri, S., 2023. FPN-SE-ResNet model for accurate diagnosis of kidney tumors using CT images. Applied Sciences, 13(17), p.9802.
23. [23] Faustine, A., Pereira, L., Bousbiat, H. and Kulkarni, S., 2020, November. UNet-NILM: A deep neural network for multi-tasks appliances state detection and power estimation in NILM. In Proceedings of the 5th International Workshop on Non-Intrusive Load Monitoring (pp. 84-88).
24. [24] Virtsionis Gkalinikis, N., Nalmpantis, C. and Vrakas, D., 2023. Variational regression for multi-target energy disaggregation. Sensors, 23(4), p.2051.
25. [25] Iandola, F.N., 2016. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size. arXiv preprint arXiv:1602.07360.
26. [26] Zhang, W., Li, J. and Qiu, X., 2019, December. SAR image superresolution using deep residual SqueezeNet. In Proceedings of the International Conference on Artificial Intelligence, Information Processing and Cloud Computing (pp. 1-5).
27. [27] Vasu, P.K.A., Gabriel, J., Zhu, J., Tuzel, O. and Ranjan, A., 2023. FastViT: A fast hybrid vision transformer using structural reparameterization. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 5785-5795).
28. [28] Talab, M.A., Awang, S. and Ansari, M.D., 2020. A Novel Statistical Feature Analysis‐Based Global and Local Method for Face Recognition. International Journal of Optics, 2020(1), p.4967034.
29. [29] Huang, G., Liu, Z., Van Der Maaten, L. and Weinberger, K.Q., 2017. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4700-4708).
30. [30] Li, Y., Wang, J., Dai, X., Wang, L., Yeh, C.C.M., Zheng, Y., Zhang, W. and Ma, K.L., 2023. How does attention work in vision transformers? A visual analytics attempt. IEEE transactions on visualization and computer graphics, 29(6), pp.2888-2900.
31. [31] Weng, O., Marcano, G., Loncar, V., Khodamoradi, A., Sheybani, N., Meza, A., Koushanfar, F., Denolf, K., Duarte, J.M. and Kastner, R., 2024. Tailor: Altering skip connections for resource-efficient inference. ACM Transactions on Reconfigurable Technology and Systems, 17(1), pp.1-23.
32. [32] Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D. and Batra, D., 2017. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision (pp. 618-626).