Optimization of MR Image Quality Using Fuzzy Logic
Abstract
Magnetic resonance image quality is a complex process involving multiple acquisition parameters such as the image matrix, number of excitations (NEX), field of view (FOV), and signal-to-noise ratio (SNR). The nonlinear relationships among these parameters can lead to unwanted artifacts in the images and negatively affect critical factors in healthcare, such as speed and accuracy. Therefore, improving MR image quality is expected to be more effectively achieved by combining these nonlinear parameters with artificial intelligence techniques, specifically fuzzy logic, rather than relying solely on traditional mathematical optimization methods, allowing for a more flexible and robust relational model. As a result of the fuzzy logic model developed for MRI image quality optimization, it has been demonstrated that fuzzy logic is an effective method for modeling nonlinear parameters. Accordingly, it has been revealed that high SNR and large matrix values among the input parameters play a direct role in image quality.
Keywords
Thanks
References
- [1] H. Semat and J. R. Albright, Introduction to Atomic and Nuclear Physics, 5th ed. New York, NY, USA: Springer, 2012, doi: 10.1007/978-1-4615-9701-8.
- [2] M. R. Madiyeva, D. T. Raissov, A. G. Kuanysheva, A. V. Rakhimbekov, M. N. Baizakova, and K. A. Tussupzhanova, “History and development prospects of magnetic resonance imaging,” Nauka i Zdravookhranenie, Vol. 20, No. 6, pp. 169–175, 2018.
- [3] J. M. Wardlaw et al., “A systematic review of the utility of 1.5 versus 3 Tesla magnetic resonance brain imaging in clinical practice and research,” European Radiology, Vol. 22, No. 11, pp. 2295–2303, 2012, doi: 10.1007/s00330-012-2500-8.
- [4] M. K. Jha, “Handling uncertain information with fuzzy logic in locating new infrastructure,” Journal of Data Science and Intelligent Systems, pp. 1–10, 2025, doi: 10.47852/bonviewJDSIS52025076.
- [5] N. R. Jagannathan, “Potential of magnetic resonance (MR) methods in clinical cancer research,” in Biomedical Translational Research: Technologies for Improving Healthcare. Singapore: Springer Nature Singapore, 2022, pp. 339–360, doi: 10.1007/978-981-16-4345-3_21.
- [6] H. Kabasawa, “MR imaging in the 21st century: Technical innovation over the first two decades,” Magnetic Resonance in Medical Sciences, Vol. 21, No. 1, pp. 71–82, 2022, doi: 10.2463/mrms.rev.2021-0011.
- [7] H. Yang et al., “Artificial intelligence for neuro MRI acquisition: A review,” Magnetic Resonance Materials in Physics, Biology and Medicine, Vol. 37, No. 3, pp. 383–396, 2024, doi: 10.1007/s10334-024-01182-7.
- [8] D. S. Lakomy et al., “Clinical implementation and initial experience with a 1.5 Tesla MR-Linac for MR-guided radiation therapy for gynecologic cancer: An R-IDEAL stage 1 and 2a first in humans feasibility study of new technology implementation,” Practical Radiation Oncology, Vol. 12, No. 4, pp. e296–e305, 2022, doi: 10.1016/j.prro.2022.03.002.
- [9] F. A. Cardillo, F. Masulli, and S. Rovetta, “Automatic approaches for CE-MRI examination of the breast: A survey,” in Proc. 2017 IEEE Int. Conf. Internet of Things (iThings), IEEE Green Computing and Communications (GreenCom), IEEE Cyber, Physical and Social Computing (CPSCom), and IEEE Smart Data (SmartData), Exeter, UK, 2017, pp. 147–154, doi: 10.1109/iThings-GreenCom-CPSCom-SmartData.2017.27.
- [10] Y.-C. Heo, H.-K. Lee, and J.-H. Cho, “Effects of NEX on SNR and artifacts in parallel MR images acquired using reference scan,” Journal of Magnetics, Vol. 18, No. 4, pp. 422–427, 2013, doi: 10.4283/JMAG.2013.18.4.422.