Feature Based Cancer Detection from QIN Breast DCE-MRI Images

Abstract

A single tumor can spread to other locations in the body in an incredibly short time frame of time, often much faster than a number of separate cases can emerge from the single tissue growths. While most doctors believe that MRI is the most widely used imaging technique for detecting breast cancer, there are other alternatives, such as ultrasound and mammography. those patients with previous tests of an evaluations as well as their treatment details were followed up to see whether the previous scans and treatment gave adequate predictions and results were later checked online (QIN, pre-breast images). Extracting feature values from images using the MATLAB function is carried out in a MATLAB macro. Based on the data gathered from the experiment, it, it will be a fairly simple procedure to pinpoint the most important and less important locations Once it is done then it is easy to use MRI segmentation with the Image tool in MATLAB. When developing the methodology, you would need to have in mind the following parameters: Mean Area, Entropy, Mean Absolute Error, Mean Square Error, Peak Signal to Noise Ratio, Standard deviation and Similarity Index were taken into account. It has been shown that the use of function parameters yielded the most efficient results with a minimum of effort and precision.

Introduction

The study focuses on improving breast cancer detection using DCE-MRI (Dynamic Contrast-Enhanced MRI) and fuzzy logic-based image processing. Breast cancer is highly prevalent, and Neoadjuvant Therapy (NAT) is used for locally aggressive tumors to improve post-surgical outcomes. Imaging techniques like MRI and ultrasound are critical, with ultrasound showing higher sensitivity (94–99%) in dense tissues compared to standard MRI (75–85%). Advanced DCE-MRI allows better tumor characterization using pharmacokinetic analysis.

Methodology:

• Public QIN DCE-MRI breast dataset from TCIA was used.

• Expert radiologists segmented tumor regions in images before and after treatment.

• Fuzzy logic (FLS) was applied for image processing, edge detection, and segmentation without requiring fixed thresholds.

• Feature extraction included geometric and texture-based metrics: Mean Area, Entropy, Standard Deviation, Pixel-Signal Noise Ratio (PSNR), Mean Absolute Error (MAE), Mean Square Error (MSE), and Dice Similarity Index.

Results:

• The Fuzzy Inference System (FIS) converted pixel features into membership values and used rules to classify tumor regions.

• The method improved image quality, achieving higher PSNR and lower MSE, with better segmentation as measured by Dice similarity.

• Example metrics from five sample images showed effective tumor detection with low error and high similarity to expert-segmented regions.

Conclusion

In this study, a Fuzzy Logic–based image processing system was successfully designed and implemented for the detection and analysis of breast cancer using QIN Breast DCE-MRI datasets. The proposed method effectively integrates Fuzzy Inference Systems (FIS) with feature extraction techniques to enhance tumor segmentation and diagnostic accuracy. The fuzzy-based segmentation technique overcomes the limitations of traditional threshold-based methods by handling image uncertainty and noise more efficiently. From the obtained results, key image quality parameters such as Mean Area, Entropy, Standard Deviation, PSNR, MAE, MSE, and Dice Similarity Index were computed to evaluate the performance of the system. Among the tested samples, Image 4 demonstrated the best performance, exhibiting the highest PSNR (0.1808 dB) and the lowest MSE (266.1686), indicating superior segmentation quality and reduced error levels. The results confirm that the proposed approach produces clearer, more reliable tumor delineation with improved similarity to the ground truth. Overall, the study demonstrates that integrating fuzzy logic with DCE-MRI feature analysis provides an efficient and intelligent framework for breast cancer detection. This technique enhances image segmentation accuracy, reduces computational errors, and offers robust performance even under varying noise conditions. Future work may extend this approach by incorporating machine learning classifiers or deep learning frameworks for automated diagnosis and larger dataset validation to further improve diagnostic reliability and clinical applicability.

References

[1] K. U. Park et al., “Neoadjuvant endocrine therapy use in early stage breast cancer during the covid-19 pandemic,” Breast Cancer Res. Treat., pp. 1–10, 2021. [2] X. Sun et al., “Neoadjuvant therapy and sentinel lymph node biopsy in HER2-positive breast cancer patients: results from the PEONY trial,” Breast Cancer Res. Treat., pp. 1–6, 2020. [3] T. Huzarski et al., “Screening with magnetic resonance imaging, mammography and ultrasound in women at average and intermediate risk of breast cancer,” Hered. Cancer Clin. Pract., vol. 15, no. 1, pp. 1–8, 2017. [4] D. M. Bardo, D. R. Biyyam, M. C. Patel, K. Wong, D. Van Tassel, and R. K. Robison, “Magnetic resonance imaging of the pediatric mediastinum,” Pediatr. Radiol., vol. 48, no. 9, pp. 1209–1222, 2018. [5] T. P. Siegel, G. Hamm, J. Bunch, J. Cappell, J. S. Fletcher, and K. Schwamborn, “Mass spectrometry imaging and integration with other imaging modalities for greater molecular understanding of biological tissues,” Mol. Imaging Biol., vol. 20, no. 6, pp. 888–901, 2018. [6] D. A. Ragab, M. Sharkas, S. Marshall, and J. Ren, “Breast cancer detection using deep convolutional neural networks and support vector machines,” PeerJ, vol. 7, p. e6201, 2019. [7] S. Punitha, A. Amuthan, and K. S. Joseph, “Benign and malignant breast cancer segmentation using optimized region growing technique,” Future Comput. Inform. J., vol. 3, no. 2, pp. 348–358, 2018. [8] V. Vaishnavi and M. Suresh, “Applications of Fuzzy Logic Approach for Assessment,” in Advances in Materials Research, Springer, 2021, pp. 1191–1198. [9] A. Mardani et al., “Application of decision making and fuzzy sets theory to evaluate the healthcare and medical problems: a review of three decades of research with recent developments,” Expert Syst. Appl., vol. 137, pp. 202–231, 2019. [10] J. Greeda, A. Mageswari, and R. Nithya, “A study on fuzzy logic and its applications in medicine,” Int. J. Pure Appl. Math., vol. 119, no. 16, pp. 1515–1525, 2018. [11] P. Hilletofth, M. Sequeira, and A. Adlemo, “Three novel fuzzy logic concepts applied to reshoring decision-making,” Expert Syst. Appl., vol. 126, pp. 133–143, 2019. [12] D. N. Utama and U. Taryana, “Fuzzy logic for simply prioritizing information in academic information system,” Int. J. Mech. Eng. Technol., vol. 10, no. 2, pp. 1594–1602, 2019. [13] C. Challoumis, “Multiple Axiomatics Method and the Fuzzy Logic,” Available SSRN 3224425, 2018. [14] G. Kaiser, “Mathematical modelling and applications in education,” Encycl. Math. Educ., pp. 553–561, 2020. [15] F. Gunawan, G. Wang, D. N. Utama, and S. Komsiyah, “Decision Support Model for Supplier Selection Using Fuzzy Logic Concept,” in 2018 International Conference on Information Management and Technology (ICIMTech), IEEE, 2018, pp. 394–399. [16] D. Sathish, S. Kamath, K. Prasad, R. Kadavigere, and R. J. Martis, “Asymmetry analysis of breast thermograms using automated segmentation and texture features,” Signal Image Video Process., vol. 11, no. 4, pp. 745–752, 2017. [17] S. Saman and S. J. Narayanan, “Survey on brain tumor segmentation and feature extraction of MR images,” Int. J. Multimed. Inf. Retr., vol. 8, no. 2, pp. 79–99, 2019. [18] H. M. Whitney, H. Li, Y. Ji, P. Liu, and M. L. Giger, “Comparison of Breast MRI Tumor Classification Using Human-Engineered Radiomics, Transfer Learning From Deep Convolutional Neural Networks, and Fusion Methods,” Proc. IEEE, vol. 108, no. 1, pp. 163–177, 2019. [19] Q. Liu, Z. Liu, S. Yong, K. Jia, and N. Razmjooy, “Computer-aided breast cancer diagnosis based on image segmentation and interval analysis,” Automatika, vol. 61, no. 3, pp. 496–506, 2020. [20] A. Ibrahim, S. Mohammed, H. A. Ali, and S. E. Hussein, “Breast cancer segmentation from thermal images based on chaotic Salp swarm algorithm,” IEEE Access, vol. 8, pp. 122121–122134, 2020. [21] B. Z. Dashevsky et al., “Breast implant-associated anaplastic large cell lymphoma: Clinical and imaging findings at a large US cancer center,” Breast J., vol. 25, no. 1, pp. 69–74, 2019. [22] F. Sadoughi, Z. Kazemy, F. Hamedan, L. Owji, M. Rahmanikatigari, and T. T. Azadboni, “Artificial intelligence methods for the diagnosis of breast cancer by image processing: a review,” Breast Cancer Targets Ther., vol. 10, p. 219, 2018. [23] M. Fayaz, I. Ullah, and D. Kim, “An optimized fuzzy logic control model based on a strategy for the learning of membership functions in an indoor environment,” Electronics, vol. 8, no. 2, p. 132, 2019. [24] Z. Ashraf, M. L. Roy, P. K. Muhuri, and Q. D. Lohani, “Interval type-2 fuzzy logic system-based similarity evaluation for image steganography,” Heliyon, vol. 6, no. 5, p. e03771, 2020.

Copyright

Copyright © 2025 Manushri Deva, Sai Prudhvi Karri. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.