Impact of Speckle Reduction Filters on Machine Learning-Based Detection of Polycystic Ovary Syndrome from Ovarian Ultrasound Images
DOI:
https://doi.org/10.29207/resti.v10i4.7658Keywords:
follicle segmentation, image preprocessing, kuan filter, machine learning, polycystic ovary syndrome, ultrasound speckle noiseAbstract
Polycystic Ovary Syndrome (PCOS) is commonly assessed with ovarian ultrasonography, but speckle can conceal follicular margins and reduce the robustness of automated interpretation. Although automated PCOS studies increasingly employ machine learning, the contribution of conventional despeckling to subsequent segmentation and classification has not been examined consistently. This study compares five classical filters - Mean, Median, Lee, Frost, and Kuan - within an interpretable machine-learning pipeline for ovarian ultrasound analysis. From a public collection of 12,680 images, a balanced sample of 300 scans (150 PCOS and 150 non-PCOS) was selected. Two radiologists produced follicle annotations, and disagreements were resolved with a third expert to obtain consensus masks. Each filtered image was segmented by adaptive thresholding with morphological refinement, after which geometric and intensity descriptors were extracted. Support Vector Machine (SVM), Random Forest (RF), k-Nearest Neighbors (k-NN), and Logistic Regression (LR) were trained using a stratified 70/30 train-test split with cross-validated hyperparameter tuning. The Kuan-LR configuration yielded the strongest result, reaching 94.44% accuracy and an AUC of 0.98, together with the best edge-preservation score and segmentation agreement. The results indicate that preprocessing materially affects the reliability of an interpretable PCOS detection pipeline and provide quantitative guidance for selecting a speckle-reduction strategy before segmentation and classification.
Downloads
References
[1] R. Azziz, “Polycystic ovary syndrome,” Nat. Rev. Dis. Primer, vol. 2, no. 1, p. 16057, 2016, doi: 10.1038/nrdp.2016.57.
[2] G. Bozdag, S. Mumusoglu, D. Zengin, E. Karabulut, and B. O. Yildiz, “The prevalence and phenotypic features of polycystic ovary syndrome: a systematic review and meta-analysis,” Hum. Reprod., vol. 31, no. 12, pp. 2841–2855, 2016, doi: 10.1093/humrep/dew218.
[3] The Rotterdam ESHRE/ASRM-sponsored PCOS consensus workshop group, “Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS),” Hum. Reprod., vol. 19, no. 1, pp. 41–47, 2004, doi: 10.1093/humrep/deh098.
[4] H. J. Teede, “Recommendations from the international evidence-based guideline for the assessment and management of polycystic ovary syndrome,” Fertil. Steril., vol. 110, no. 3, pp. 364–379, 2018, doi: 10.1016/j.fertnstert.2018.05.004.
[5] M. Li, “SoK: Intelligent Detection for Polycystic Ovary Syndrome (PCOS),” medRxiv, 2024, doi: 10.1101/2024.12.25.24319623.
[6] D. Dewailly, “The physiology and clinical utility of anti-Müllerian hormone in women,” Hum. Reprod. Update, vol. 20, no. 3, pp. 370–385, 2014, doi: 10.1093/humupd/dmt062.
[7] S. Hartati, A. Musdholifah, P. D. Wulaning Ayu, and J. Dasuki, “Follicle Detection Model on Ovarian Ultrasound Image,” presented at the 2022 Seventh International Conference on Informatics and Computing (ICIC), Denpasar, Bali, Indonesia: IEEE, 2022, pp. 1–8. doi: 10.1109/ICIC56845.2022.10006915.
[8] D. T. Kuan, A. A. Sawchuk, T. C. Strand, and P. Chavel, “Adaptive Noise Smoothing Filter for Images with Signal-Dependent Noise,” IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-7, no. 2, pp. 165–177, 1985, doi: 10.1109/TPAMI.1985.4767641.
[9] R. F. Wagner, S. W. Smith, J. M. Sandrik, and H. Lopez, “Statistics of Speckle in Ultrasound B-Scans,” IEEE Trans. Sonics Ultrason., vol. 30, no. 3, pp. 156–163, 1983, doi: 10.1109/T-SU.1983.31404.
[10] N. Thakur, N. U. Khan, and S. D. Sharma, “A Two Phase Ultrasound Image De-speckling Framework by Nonlocal Means on Anisotropic Diffused Image Data,” Informatica, vol. 47, no. 2, 2023, doi: 10.31449/inf.v47i2.4378.
[11] W. Zhang, G. Peng, C. Li, Y. Chen, and Z. Zhang, “A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals,” Sensors, vol. 17, no. 2, p. 425, 2017, doi: 10.3390/s17020425.
[12] A. Achim, A. Bezerianos, and P. Tsakalides, “Novel Bayesian multiscale method for speckle removal in medical ultrasound images,” IEEE Trans. Med. Imaging, vol. 20, no. 8, pp. 772–783, 2001, doi: 10.1109/42.938245.
[13] J.-S. Lee, “Digital Image Enhancement and Noise Filtering by Use of Local Statistics,” IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-2, no. 2, pp. 165–168, 1980, doi: 10.1109/TPAMI.1980.4766994.
[14] N. A. Ali and M. M. Abd El-kareem, “A Dual-Metric Framework for Image Filtering Performance Assessment via Eigenvalue Variance and Restoration Error Analysis,” 2025, doi: 10.21203/rs.3.rs-7474908/v1.
[15] F. Laputin, I. Sidorov, and A. Moshkin, “Computer Vision Methods for Assessing Ovarian Reserve,” presented at the 2024 International Russian Smart Industry Conference (SmartIndustryCon), Sochi, Russian Federation: IEEE, 2024, pp. 656–661. doi: 10.1109/SmartIndustryCon61328.2024.10515730.
[16] D. Bafila, A. Verma, and B. Singh, “Mathematically Driven Identification and Generalizable Optimization of Smoothing Parameters for Speckle Noise Reduction in Ultrasound Imaging,” J. Imaging Inform. Med., 2026, doi: 10.1007/s10278-025-01815-w.
[17] A. Alamoudi, “A Deep Learning Fusion Approach to Diagnosis the Polycystic Ovary Syndrome (PCOS),” Appl. Comput. Intell. Soft Comput., vol. 2023, pp. 1–15, 2023, doi: 10.1155/2023/9686697.
[18] V. S. Frost, J. A. Stiles, K. S. Shanmugan, and J. C. Holtzman, “A Model for Radar Images and Its Application to Adaptive Digital Filtering of Multiplicative Noise,” IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-4, no. 2, pp. 157–166, 1982, doi: 10.1109/TPAMI.1982.4767223.
[19] M. Li, “Intelligent detection for Polycystic Ovary Syndrome (PCOS): Taxonomy, datasets and detection tools,” Comput. Struct. Biotechnol. J., vol. 27, pp. 1578–1599, 2025, doi: 10.1016/j.csbj.2025.04.011.
[20] J. Harikiran, E. Vamsidhar, B. Srinivasa Rao, and B. Saichandana, “Follicle Detection in Ultrasound Images using Adaptive Clustering Algorithms and Empirical Mode Decomposition,” Int. J. Innov. Technol. Explor. Eng., vol. 9, no. 2, pp. 2137–2141, 2019, doi: 10.35940/ijitee.B7684.129219.
[21] D. Saha, A. Mandal, and R. Ghosh, “MU Net: Ovarian Follicle Segmentation Using Modified U-Net Architecture,” Int. J. Eng. Adv. Technol., vol. 11, no. 4, pp. 30–35, 2022, doi: 10.35940/ijeat.D3419.0411422.
[22] V. K. Singh, “HaTU-Net: Harmonic Attention Network for Automated Ovarian Ultrasound Quantification in Assisted Pregnancy,” Diagnostics, vol. 12, no. 12, p. 3213, 2022, doi: 10.3390/diagnostics12123213.
[23] P. Singh, M. Diwakar, V. Singh, S. Kadry, and J. Kim, “A new local structural similarity fusion-based thresholding method for homomorphic ultrasound image despeckling in NSCT domain,” J. King Saud Univ. - Comput. Inf. Sci., vol. 35, no. 7, p. 101607, 2023, doi: 10.1016/j.jksuci.2023.101607.
[24] Adiwijaya, “Follicle detection on the USG images to support determination of polycystic ovary syndrome,” J. Phys. Conf. Ser., vol. 622, no. 1, p. 012027, 2015, doi: 10.1088/1742-6596/622/1/012027.
[25] U. N. Wisesty, J. Nasri, and Adiwijaya, “Modified Backpropagation Algorithm for Polycystic Ovary Syndrome Detection Based on Ultrasound Images,” in Soft Computing and Data Mining: Proceedings of the Second International Conference on Soft Computing and Data Mining (SCDM 2016), vol. 549, in Advances in Intelligent Systems and Computing, vol. 549. , Cham, Switzerland: Springer, 2017, pp. 141–151. doi: 10.1007/978-3-319-51281-5_15.
[26] A. Indirani, “PCOS Dataset.” figshare, 2024. doi: 10.6084/M9.FIGSHARE.27682557.V1.
[27] P. Afshari, “Refraction Based Elevational Angular Compounding for Ultrasound Speckle Reduction,” Technische Universität München, Munich, Germany, 2024. [Online]. Available: https://mediatum.ub.tum.de/1739914
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright in each article belongs to the author
- The author acknowledges that the RESTI Journal (System Engineering and Information Technology) is the first publisher to publish with a license Creative Commons Attribution 4.0 International License.
- Authors can enter writing separately, arrange the non-exclusive distribution of manuscripts that have been published in this journal into other versions (eg sent to the author's institutional repository, publication in a book, etc.), by acknowledging that the manuscript has been published for the first time in the RESTI (Rekayasa Sistem dan Teknologi Informasi) journal ;
