DEVELOPMENT AND VALIDATION OF A PROTOTYPE PIPELINE FOR BIOMEDICAL IMAGE ANALYSIS IN CARDIOVASCULAR MONITORING
https://doi.org/10.53360/2788-7995-2026-2(22)-7
Abstract
Cardiovascular diseases remain the leading cause of mortality, which determines the relevance of developing robust methods for cardiac image analysis and patient monitoring. This paper presents and validates a prototype pipeline for automated preprocessing and segmentation of biomedical images, developed within the framework of a scientific project (IRN: AP05132044). The experimental protocol is implemented on the Heart Database dataset (18 patients, 3D+t MRI (three-dimensional with time dimension)), using expert endocardial masks and 36 slices (diastole/systole, 2 slices per patient). Performance evaluation was conducted using PSNR, SSIM, Dice, and IoU metrics. A comparative analysis was performed across five preprocessing modes: none, Gaussian, wavelet, NLM, and hybrid (wavelet + NLM + CLAHE). The results show that the NLM method achieves the best denoising performance (PSNR = 26.07±0.33 dB vs. 22.85±0.21 dB for the baseline, p = 1.46×10⁻¹¹, Wilcoxon test), while the highest segmentation accuracy is obtained with hybrid preprocessing (Dice = 0.681±0.176 vs. 0.636±0.153 without preprocessing, p = 0.018). An analysis of challenging cases revealed that segmentation performance is strongly influenced by endocardial contrast and geometric complexity across different phases of the cardiac cycle. The obtained results demonstrate a statistically significant impact of the preprocessing stage on downstream segmentation quality and confirm the feasibility of the integrated “preprocessing-segmentation-evaluation” pipeline as an engineering foundation for clinical frameworks. The proposed approach is designed for reproducible performance under real-world data heterogeneity and can serve as a basis for further integration into clinical information systems.
Keywords
biomedical images,
cardiac monitoring,
deep learning,
heart segmentation,
denoising,
ejection fraction,
remote healthcare
Supporting agencies: This research has been funded by the Science Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan (grant: №AP25794129 «Development of an Algorithm for Filtering and Preprocessing Biomedical Images for Cardiology Diagnostics»).
About the Authors
M. Kabdullin
Satbayev University
Kazakhstan
Kazakhstan
Maxat Kabdullin,
050013, Almaty 22 Satbayev Street
Лязат Найзабаева
International IT University
Kazakhstan
Kazakhstan
Lyazat Naizabayeva – Doctor of Technical Sciences, Research Professor, Department of Information Systems,
050040, Almaty 34/1 Manasa Street
S. A.R. Al-Haddad
Universiti Putra Malaysia
Malaysia
Malaysia
Syed Abdul Rahman Al-Haddad – Professor, Department of Computer and Communication Systems Engineering, Faculty of Engineering,
Selangor, 43400 UPM Serdang
A. Kabdullin
Satbayev University
Kazakhstan
Kazakhstan
Azat Kabdullin – Research Lecturer, Department of Information Systems,
050013, Almaty 22 Satbayev Street
Asiya Issayeva
National Hospital of the Medical Center of the President's Affairs Administration of the Republic of Kazakhstan
Kazakhstan
Kazakhstan
Asiya Issayeva – Head of the Center for Innovation and Education,
050000, Almaty 16 Baiseitova Street
References
1. World Health Organization. Cardiovascular diseases (CVDs), 2025. Available: https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds).
2. A survey on deep learning in medical image analysis / G. Litjens et al // Medical Image Analysis. – 2017. – Vol. 42. – P. 60-88. https://doi.org/10.1016/j.media.2017.07.005.
3. Shen D. Deep Learning in Medical Image Analysis / D. Shen, G. Wu, H.-I. Suk // Annual Review of Biomedical Engineering. – 2017. – Vol. 19. – P. 221-248. https://doi.org/10.1146/annurev-bioeng071516-044442.
4. A guide to deep learning in healthcare / A. Esteva et al // Nature Medicine. – 2019. – Vol. 25. – P. 24-29. https://doi.org/10.1038/s41591-018-0316-z.
5. Ronneberger O. U-Net: Convolutional Networks for Biomedical Image Segmentation / O. Ronneberger, P. Fischer, T. Brox // MICCAI. – 2015. – Р. 234-241. https://arxiv.org/abs/1505.04597 (accessed: 25.03.2026).
6. UNet++: A Nested U-Net Architecture for Medical Image Segmentation / Z. Zhou et al // DLMIA/ML-CDS. – 2018. – Р. 3-11. https://doi.org/10.1007/978-3-030-00889-5_1.
7. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation / F. Isensee et al // Nature Methods. – 2021. – vol. 18. – Р. 203-211. https://doi.org/10.1038/s41592-020-01008-z.
8. Segment Anything / A. Kirillov et al // ICCV. – 2023. – Р. 4015-4026. Available: https://openaccess.thecvf.com/content/ICCV2023/html/Kirillov_Segment_Anything_ICCV_2023_paper.html
9. Segment anything in medical images / J. Ma et al // Nature Communications. – 2024. – vol. 15, Art. 654. https://doi.org/10.1038/s41467-024-44824-z.
10. Video-based AI for beat-to-beat assessment of cardiac function / D. Ouyang et al // Nature. – 2020. – vol. 580. – Р. 252-256. https://doi.org/10.1038/s41586-020-2145-8.
11. Deep Learning for Segmentation Using an Open Large-Scale Dataset in 2D Echocardiography / S. Leclerc et al // IEEE Transactions on Medical Imaging. – 2019. – vol. 38, № 9. – Р. 2198-2210. https://doi.org/10.1109/TMI.2019.2900516.
12. Topol E.J. High-performance medicine: the convergence of human and artificial intelligence / E.J. Topol // Nature Medicine. – 2019. – vol. 25. – Р. 44-56. https://doi.org/10.1038/s41591-018-0300-7.
13. Secure, privacy-preserving and federated machine learning in medical imaging / G.A. Kaissis et al // Nature Machine Intelligence. – 2020. – vol. 2. – Р. 305-311. https://doi.org/10.1038/s42256-020- 0186-1.
14. Holzinger A. Measuring the Quality of Explanations: The System Causability Scale (SCS) / A. Holzinger, A. Carrington, H. Müller // KI – Künstliche Intelligenz. – 2020. – vol. 34. – Р. 193-198. https://doi.org/10.1007/s13218-020-00636-z.
15. Tjoa E. A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI / E. Tjoa, C. Guan // Information Fusion. – 2021. – vol. 79. – Р. 263-289. https://doi.org/10.1016/j.inffus.2021.10.006.
16. Kabdullin A. Optimizing Neural Network Performance to Predict Coronary Heart Disease, Proc. / A. Kabdullin, M. Kabdullin, L. Naizabayeva // IEEE SIST. – 2021. – Р. 1-4. https://doi.org/10.1109/SIST50301.2021.9465925.
17. Multi-Centre, Multi-Vendor & Multi-Disease Cardiac Image Segmentation: The M&Ms Challenge / V.M. Campello et al // IEEE Transactions on Medical Imaging. – 2021. – vol. 40, № 12. – Р. 3543- 3554. https://doi.org/10.1109/TMI.2021.3090082.
18. UNETR: Transformers for 3D Medical Image Segmentation, Proc. / A. Hatamizadeh et al // WACV. – 2022. – Р. 574-584. https://doi.org/10.1109/WACV51458.2022.00181.
19. Image denoising by sparse 3-D transform-domain collaborative filtering / K. Dabov et al // IEEE Transactions on Image Processing. – 2007. – vol. 16, № 8. – Р. 2080-2095. https://doi.org/10.1109/TIP.2007.901238.
20. Noise2Noise: Learning Image Restoration without Clean Data / J. Lehtinen et al // ICML. – 2018. – Р. 2965-2974. Available: https://proceedings.mlr.press/v80/lehtinen18a/lehtinen18a.pdf (accessed: 25.03.2026).
21. Krull A. Noise2Void – Learning Denoising From Single Noisy Images / A. Krull, T.-O. Buchholz, F. Jug // CVPR. – 2019. – Р. 2129-2137. Available: https://openaccess.thecvf.com/content_CVPR_2019/html/Krull_Noise2Void_-_Learning_Denoising_From_Single_Noisy_Images_CVPR_2019_paper.html (accessed: 25.03.2026).
22. The Heart Database (3D+t cardiac MRI with expert and automated segmentations) / J. Cousty et al // Available: https://perso.esiee.fr/~dpt-it/a2si/heart/data/HeartDatabase.tgz (accessed: 25.03.2026).
Review
For citations:
Kabdullin M., Найзабаева Л., Al-Haddad S.A., Kabdullin A., Issayeva A. DEVELOPMENT AND VALIDATION OF A PROTOTYPE PIPELINE FOR BIOMEDICAL IMAGE ANALYSIS IN CARDIOVASCULAR MONITORING. Bulletin of Shakarim University. Technical Sciences. 2026;(2(22)):71-80. https://doi.org/10.53360/2788-7995-2026-2(22)-7
Views:
6
JATS XML
Email this article 
