APPLICATION OF EXPLAINABLE ARTIFICIAL INTELLIGENCE METHODS FOR SEGMENTATION AND CLASSIFICATION OF ATHEROSCLEROTIC PLAQUES FROM OCT DATA

Highlights

This study presents and validates a hybrid framework that combines deep learning with Explainable Artificial Intelligence (XAI) for the segmentation and classification of atherosclerotic plaques. The incorporation of XAI techniques improves the interpretability of deep learning models, thereby facilitating their translation into clinical practice for the diagnosis of atherosclerosis. The results highlight the potential of XAI-enabled approaches to support the development of reliable, transparent, and clinically applicable decision support systems in cardiology.

Aim. To develop and validate a comprehensive method for automated segmentation and classification of morphological features of atherosclerotic plaques based on optical coherence tomography using explainable artificial intelligence approaches.

Methods. Nine state-of-the-art segmentation neural network architectures were trained on a dataset of 103 OCT studies annotated for four key plaque features. To assess interpretability, nine XAI methods were applied.

Results. The best segmentation results were achieved with DeepLabV3+ and MA-Net models (DSC up to 72.1%). LayerCAM and HiResCAM methods provided the most stable and accurate visualization of attention areas.

Conclusion. The proposed approach demonstrates high segmentation accuracy and interpretability, enhancing clinical trust and paving the way for future development of explainable AI systems in cardiology.

1. F. Zibaeenejad, S. S. Mohammadi, M. Sayadi, F. Safari, and M. J. Zibaeenezhad, “Ten-year atherosclerosis cardiovascular disease (ASCVD) risk score and its components among an Iranian population: a cohort-based cross-sectional study,” BMC Cardiovascular Disorders, vol. 22, no. 1, p. 162, Mar. 2022, doi: 10.1186/s12872-022-02601-0.

2. N. A. Kochergin, A. M. Kochergina, V. I. Ganyukov, and O. L. Barbarash, “Vulnerable Plaques in Patients with Stable Coronary Artery Disease,” in Horizons in World Cardiovascular Research, vol. 21, E. H. Bennington, Ed., Nova Science Publishers, 2021, pp. 187–203.

3. M. Vaduganathan, G. A. Mensah, J. V. Turco, V. Fuster, and G. A. Roth, “The Global Burden of Cardiovascular Diseases and Risk,” Journal of the American College of Cardiology, vol. 80, no. 25, pp. 2361–2371, Mar. 2022, doi: 10.1016/j.jacc.2022.11.005.

4. M. Araki et al., “Predictors of Rapid Plaque Progression,” JACC: Cardiovascular Imaging, vol. 14, no. 8, pp. 1628–1638, Mar. 2021, doi: 10.1016/j.jcmg.2020.08.014.

5. P. Baruś et al., “Comparative Appraisal of Intravascular Ultrasound and Optical Coherence Tomography in Invasive Coronary Imaging: 2022 Update,” Journal of Clinical Medicine, vol. 11, no. 14, p. 4055, Mar. 2022, doi: 10.3390/jcm11144055.

6. N. A. Kochergin, K. A. M. Kochergin, “Intravascular Ultrasound as a Method for Detecting Unstable Atherosclerotic Plaques in Coronary Arteries (Literature Review),” Medical Imaging, vol. 4, pp. 82–87, 2017, doi: 10.24835/1607-0763-2017-4-82-87.

7. N. A. Kochergin and A. M. Kochergina, “Capabilities of Optical Coherence Tomography and Intravascular Ultrasound in the Detection of Unstable Plaques in Coronary Arteries,” Cardiovascular Therapy and Prevention, vol. 21, no. 1, p. 2909, Jan. 2022, doi: 10.15829/1728-8800-2022-2909.

8. D. Chamié et al., “Optical Coherence Tomography Versus Intravascular Ultrasound and Angiography to Guide Percutaneous Coronary Interventions,” Circulation: Cardiovascular Interventions, vol. 14, no. 3, p. e009452, Mar. 2021, doi: 10.1161/CIRCINTERVENTIONS.120.009452.

9. N.A. Kochergin, A.M. Kochergina, V.I. Ganjukov, O.L. Barbarash. Vulnerable atherosclerotic plaques of coronary arteries in patients with stable coronary artery disease. Complex Issues of Cardiovascular Diseases. 2018; 7 (3): 65-71. DOI: 10.17802/2306-1278-2018-7-3-65-71

10. L. S. Athanasiou et al., “Methodology for Fully Automated Segmentation and Plaque Characterization in Intracoronary Optical Coherence Tomography Images,” Journal of Biomedical Optics, vol. 19, no. 2, p. 026009, Mar. 2014, doi: 10.1117/1.JBO.19.2.026009.

11. M. C. Williams et al., “Artificial Intelligence and Machine Learning for Cardiovascular Computed Tomography (CCT): A White Paper of the Society of Cardiovascular Computed Tomography (SCCT),” Journal of Cardiovascular Computed Tomography, vol. 0, no. 0, Mar. 2024, doi: 10.1016/j.jcct.2024.08.003.

12. C. Kolluru, D. Prabhu, Y. Gharaibeh, H. Bezerra, G. Guagliumi, and D. Wilson, “Deep Neural Networks for A-line-based Plaque Classification in Coronary Intravascular Optical Coherence Tomography Images,” Journal of Medical Imaging, vol. 5, no. 4, p. 044504, Mar. 2018, doi: 10.1117/1.JMI.5.4.044504.

13. C. Rudin, “Stop Explaining Black Box Machine Learning Models for High-Stakes Decisions and Use Interpretable Models Instead,” Nature Machine Intelligence, vol. 1, no. 5, pp. 206–215, Nov. 2018, doi: 10.1038/s42256-019-0048-x.

14. W. Samek, T. Wiegand, and K.-R. Müller, “Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models,” Aug. 2017, Accessed: Jul. 06, 2025. [Online]. Available: https://arxiv.org/pdf/1708.08296.

15. R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” International Journal of Computer Vision, vol. 128, no. 2, pp. 336–359, Oct. 2016, doi: 10.1007/s11263-019-01228-7.

16. Y. Zhang, P. Tiňo, A. Leonardis, and K. Tang, “A Survey on Neural Network Interpretability,” 2021, Accessed: Jul. 06, 2025. [Online]. Available: https://www.acm.org/media-center/2019/march/turing-award-2018.

17. S. V. Emelyanov, D. A. Makarov, “Stabilization of Helicopter Pitch Angle under Various Flight Conditions Using Coordinate-Operator and Operator Feedbacks,” Artificial Intelligence and Decision Making, no. 4, pp. 68–80, 2011. [Online]. Available: https://www.mathnet.ru/php/archive.phtml?wshow=paper&jrnid=iipr&paperid=479&option_lang=eng.

18. E. A. Vasilieva, T. V. Pavlova, “Challenges and Prospects for the Introduction of Artificial Intelligence in Medicine,” State and Municipal Management Scholar Notes, vol. 1, no. 1, pp. 25–32, Mar. 2022, doi: 10.22394/2079-1690-2022-1-1-25-32.

19. E. I. Tarlovskaya et al., “Analysis of the Influence of Background Therapy for Comorbidities Prior to Infection on the Risk of Lethal COVID Outcome. Data from the International ACTIV SARS-CoV-2 Registry (‘Analysis of Chronic Non-Infectious Diseases Dynamics after COVID-19 Infection in Adult Patients SARS-CoV-2’),” Kardiologiya, vol. 61, no. 9, pp. 20–32, 2021, doi: 10.18087/CARDIO.2021.9.N1680.

20. Supervisely, “Supervisely Computer Vision Platform,” Mar. 2023. [Online]. Available: https://supervisely.com.

21. Uncovering unstable plaques: deep learning segmentation in optical coherence tomography // Laptev V.V., Danilov V.V, Ovcharenko E.A, Klyshnikov K.Y, Kolesnikov A.Y, Arnt A.A, Bessonov I.S, Litvinyuk N.V, Kochergin N.A // Computer Optics. – 2025. – Vol. 49. –¬ № 5. – P. 775–793

22. Segmentation and quantification of atherosclerotic plaques in optical coherence tomography / V. V. Danilov, V.V. Laptev, K. Y. Klyshnikov, I. S. Bessonov [et al.] // Computers in Biology and Medicine. – 2025. – Vol. 197