JBT Journal
Advanced

Raqamli Transformatsiya va Sun’iy Intellekt

APPROACHES TO EXTRACTING INFORMATIVE FEATURES FROM MRI IMAGES USING FILTERS: A METHODOLOGICAL STUDY OF PREPROCESSING FOR BREAST CANCER IMAGE ANALYSIS

Ungboyeva Mukaddas1

1 Jizzakh branch of the National University of Uzbekistan named after Mirzo Ulugbek

Keywords

radiomic features Gaussian smoothing intensity normalization preprocessing breast MRI CLAHE Laplacian of Gaussian

Abstract

Reliable breast magnetic resonance imaging analysis begins with control of image intensity and spatial detail before features are extracted. This article examines two preprocessing methods, percentile based intensity normalization and contrast limited adaptive histogram equalization, together with two filter types, Gaussian smoothing and the Laplacian of Gaussian. Their roles are separated into intensity harmonization, local contrast redistribution, suppression of fine variation, and extraction of scale dependent structures. A reproducible two dimensional demonstration is performed on a publicly licensed published breast DCE-MRI image. Original, preprocessed, and filtered panels are presented alongside intensity distributions, line profiles, and filter sensitivity diagrams. Quantitative measurements describe changes in the displayed image rather than diagnostic performance. Gaussian filtering attenuates high frequency variation while LoG produces signed structural response maps that must be interpreted separately from anatomical grayscale images

Views: 123 Published: 2026-10-10 Pages: 63-71 Volume: 4 Issue: 5

References

  1. Z. L. Yang, “Breast dce-mri.jpg,” Wikimedia Commons, 2021. Image, 378 × 297 pixels, CC BY 4.0. Accessed Oct. 3, 2026. https://commons.wikimedia.org/wiki/File:Breast_dce-mri.jpg
  2. N. J. Tustison et al., “N4ITK: Improved N3 Bias Correction,” IEEE Transactions on Medical Imaging, vol. 29, no. 6, pp. 1310–1320, 2010. doi: 10.1109/TMI.2010.2046908. https://doi.org/10.1109/TMI.2010.2046908
  3. K. Zuiderveld, “Contrast Limited Adaptive Histogram Equalization,” in Graphics Gems IV, P. S. Heckbert, Ed. Academic Press, 1994, pp. 474–485. https://www.cse.unr.edu/~bebis/CS474/StudentPaperPresentations/1994%20-%20CLAHE.pdf
  4. The Cancer Imaging Archive, “Duke-Breast-Cancer-MRI,” dataset collection. Accessed Oct. 3, 2026. https://www.cancerimagingarchive.net/collection/duke-breast-cancer-mri/
  5. R. M. Mann et al., “Breast cancer screening in women with extremely dense breasts recommendations of the European Society of Breast Imaging (EUSOBI),” European Radiology, vol. 32, no. 6, pp. 4036–4045, 2022. doi: 10.1007/s00330-022-08617-6. https://doi.org/10.1007/s00330-022-08617-6
  6. P. Whybra et al., “The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights,” Radiology, vol. 310, no. 2, article e231319, 2024. doi: 10.1148/radiol.231319. https://doi.org/10.1148/radiol.231319
  7. Z. L. Yang et al., “Detection and Classification of Breast Lesions With Readout-Segmented Diffusion-Weighted Imaging in a Large Chinese Cohort,” Frontiers in Oncology, vol. 11, article 636471, 2021. doi: 10.3389/fonc.2021.636471. https://doi.org/10.3389/fonc.2021.636471