DPHT-Net: A Novel Self-Supervised Hybrid CNN-Transformer Approach for Automated Pulmonary Embolism Classification in CT Pulmonary Angiogram Scans

Authors

  • Abeer Abdelhamid

    Electronics and Communications Engineering Department, Faculty of Engineering, Mansoura University,
    Mansoura 35516, Egypt

    Higher Technological Institute of Applied Health Sciences, Mansoura 35511, Egypt
  • Hossam El-Din Moustafa

    Electronics and Communications Engineering Department, Faculty of Engineering, Mansoura University,
    Mansoura 35516, Egypt

    Faculty of Artificial Intelligence and Information, Horus University, New Damietta 34517, Egypt
  • Hala B. Nafea

    Electronics and Communications Engineering Department, Faculty of Engineering, Mansoura University,
    Mansoura 35516, Egypt
  • Ehab H. Abdelhay

    Electronics and Communications Engineering Department, Faculty of Engineering, Mansoura University,
    Mansoura 35516, Egypt

    Faculty of Engineering, Mansoura National University, Gamasa 35712, Egypt
  • Mohammed M. Abo-Zahhad

    Electrical Engineering Department, Faculty of Engineering, Sohag University, Sohag 82524, Egypt
  • Amir El-Ghamry

    Computer Science Department, Faculty of Computers and Information, Mansoura University, Mansoura 35516, Egypt

DOI:

https://doi.org/10.30564/jeis.v8i1.13224
Received: 28 February 2026 | Revised: 13 April 2026 | Accepted: 20 April 2026 | Published Online: 25 April 2026

Abstract

Pulmonary embolism (PE) is a life-threatening condition that requires timely and accurate diagnosis to enable appropriate treatment and reduce mortality rates. Despite significant advances in medical imaging, reliable detection of PE in computed tomography pulmonary angiography (CTPA) remains challenging due to the subtle appearance of emboli, variations in contrast, and the high-dimensional nature of volumetric data. These challenges necessitate robust and efficient automated methods to support clinical decision-making. In this study, we propose a diffusion-pretrained hybrid Convolutional Neural Network (CNN)-Transformer network (DPHT-Net), a lightweight architecture designed to effectively capture both local and global embolic patterns in CTPA scans. The proposed framework integrates self-supervised diffusion-based pretraining, a CNN-based module for local feature extraction and refinement, and a Transformer-based component for modeling long-range inter-slice dependencies. In addition, a sinh–cosh-based preprocessing step is introduced to enhance image contrast and highlight subtle embolic regions. The proposed model was evaluated on the RSNA-STR PE dataset. Experimental results demonstrate that DPHT-Net achieves an accuracy of 96.4% and an F1-score of 93.8%, outperforming conventional CNN-based methods by 11.9% in accuracy and 12.7% in F1-score, and surpassing Transformer-based approaches by 5.4% and 4.2%, respectively. These results indicate that DPHT-Net provides a robust, computationally efficient, and clinically applicable solution for automated PE detection, offering a promising direction for volumetric medical image analysis and computer-aided diagnosis systems.

Keywords:

Artificial Intelligence (AI); Computed Tomography Pulmonary Angiograph (CTPA); Diffusion Pretrained Hybrid CNN-Transformer (DPHT); Pulmonary Embolism (PE)

References

[1] Uzelac, B., Stanković, S., 2026. Artificial Intelligence-Driven Integration of ECG and Molecular Biomarkers in Pulmonary Embolism. International Journal of Molecular Sciences. 27(2), 813.

[2] Rivas, L.F., 2023. Clinical characterization of patients with venous thromboembolic disease in 2 reference centers in El Salvador. Blood. 142(Supplement 1), 5555.

[3] Nitha, V.R., Vinod Chandra, S.S., Valsalan, P., et al., 2025. A deep learning framework for the segmentation and quantitative analysis of pulmonary embolism. Engineering Applications of Artificial Intelligence. 155, 110972.

[4] Abdelhamid, A., El-Ghamry, A., Abdelhay, E.H., et al., 2025. Improved pulmonary embolism detection in CT pulmonary angiogram scans with hybrid vision transformers and deep learning techniques. Scientific Reports. 15(1), 31443.

[5] Douillet, D., Roy, P.-M., Penaloza, A., 2021. Suspected acute pulmonary embolism: gestalt, scoring systems, and artificial intelligence. Seminars in Respiratory and Critical Care Medicine. 42(2), 176–182.

[6] Mohanarajan, M., Salunke, P.P., Arif, A., et al., 2025. Advancements in Machine Learning and Artificial Intelligence in the Radiological Detection of Pulmonary Embolism. Cureus. 17(1).

[7] Williams, L.-M.S., Walker, G.R., Loewenherz, J.W., et al., 2020. Association of contrast and acute kidney injury in the critically ill: A propensity-matched study. Chest. 157(4), 866–876.

[8] Zhou, Q., Huang, R., Xiong, X., et al., 2025. Prediction of pulmonary embolism by an explainable machine learning approach in the real world. Scientific Reports. 15(1), 835.

[9] Bass, A.R., Fields, K.G., Goto, R., et al., 2017. Clinical decision rules for pulmonary embolism in hospitalized patients: A systematic literature review and meta-analysis. Thrombosis and Haemostasis. 117(11), 2176–2185.

[10] Stamate, E., Piraianu, A.-I., Ciobotaru, O.R., et al., 2024. Revolutionizing cardiology through artificial intelligence—Big data from proactive prevention to precise diagnostics and cutting-edge treatment—A comprehensive review of the past 5 years. Diagnostics. 14(11), 1103.

[11] Beam, A.L., Drazen, J.M., Kohane, I.S., et al., 2023. Artificial intelligence in medicine. New England Journal of Medicine. 388(13), 1220–1221.

[12] Rajkomar, A., Dean, J., Kohane, I., 2019. Machine learning in medicine. New England Journal of Medicine. 380(14), 1347–1358.

[13] Huhtanen, H., Nyman, M., Mohsen, T., et al., 2022. Automated detection of pulmonary embolism from CT-angiograms using deep learning. BMC Medical Imaging. 22(1), 43.

[14] Lam, B.D., Ma, S., Kovalenko, I., et al., 2025. Using a transformer language model to curate a pulmonary embolism dataset from the Medical Information Mart for Intensive Care IV: MIMIC-IV-Ext-PE. Research and Practice in Thrombosis and Haemostasis. 9(4), 102896.

[15] Omiye, J.A., Gui, H., Rezaei, S.J., et al., 2024. Large language models in medicine: the potentials and pitfalls: a narrative review. Annals of Internal Medicine. 177(2), 210–220.

[16] Abdelhaliem, I., Dixon, J., Abdelhamid, A., et al., 2025. A Multimodal Adaptive Inter-Region Attention-Guided Network for Brain Tumor Classification. IEEE Access. 13, 187964–187975.

[17] Qiao, Y., Gao, Y., Chen, Y., et al., 2025. Quantitative assessment and risk stratification of random acute pulmonary embolism cases using a deep learning model based on computed tomography pulmonary angiography images. Quantitative Imaging in Medicine and Surgery. 15(3), 1950.

[18] Condrea, F., Rapaka, S., Itu, L., et al., 2024. Anatomically aware dual-hop learning for pulmonary embolism detection in CT pulmonary angiograms. Computers in Biology and Medicine. 174, 108464.

[19] Ayobi, A., Chang, P.D., Chow, D.S., et al., 2024. Performance and clinical utility of an artificial intelligence-enabled tool for pulmonary embolism detection. Clinical Imaging. 113, 110245.

[20] Khan, M., Shah, P.M., Khan, I.A., et al., 2023. IoMT-enabled computer-aided diagnosis of pulmonary embolism from computed tomography scans using deep learning. Sensors. 23(3), 1471.

[21] Suman, S., Singh, G., Sakla, N., et al., 2021. Attention based CNN-LSTM network for pulmonary embolism prediction on chest computed tomography pulmonary angiograms. In Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Strasbourg, France, 27 September–1 October 2021; pp. 356–366.

[22] Yang, X., Lin, Y., Su, J., et al., 2019. A two-stage convolutional neural network for pulmonary embolism detection from CTPA images. IEEE Access. 7, 84849–84857.

[23] Ma, X., Ferguson, E.C., Jiang, X., et al., 2022. A multitask deep learning approach for pulmonary embolism detection and identification. Scientific Reports. 12(1), 13087.

[24] Ting, I.-H., Tseng, Y.-J., Lin, Y.-S., 2026. Application of deep learning techniques in non-contrast computed tomography pulmonary angiogram for pulmonary embolism diagnosis. arXiv preprint. arXiv:2601.00925.

[25] Biret, C.B., Gurbuz, S., Akbal, E., et al., 2025. Advancing Pulmonary Embolism Detection with Integrated Deep Learning Architectures. Journal of Imaging Informatics in Medicine. 39, 186–201.

[26] Singh, G., Singh, A., Kainth, T., et al., 2025. Comparing efficiency of an attention-based deep learning network with contemporary radiological workflow for pulmonary embolism detection on CTPA: A retrospective study. European Journal of Radiology Open. 14, 100657.

[27] da Silva, L.O., da Silva, M.C.B., Ribeiro, G.A.S., et al., 2024. Artificial intelligence-based pulmonary embolism classification: Development and validation using real-world data. Plos One. 19(8), e0305839.

[28] Huang, S.-C., Kothari, T., Banerjee, I., et al., 2020. PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging. NPJ Digital Medicine. 3(1), 61.

[29] Kiourt, C., Feretzakis, G., Dalamarinis, K., et al., 2021. Pulmonary embolism identification in computerized tomography pulmonary angiography scans with deep learning technologies in COVID-19 patients. arXiv preprint. arXiv:2105.11187.

[30] Amudha, T.K., Sunitha, R., 2025. Multi-Modal Deep Learning for Pulmonary Embolism Classification: A Late Fusion Approach Combining CTPA Imaging and EMR Data. In proceedings of the 2025 International Conference on Emerging Technologies in Engineering Applications (ICETEA), Puducherry, India, 5–6 June 2025.

[31] Zhu, H., Tao, G., Jiang, Y., et al., 2024. Automatic detection of pulmonary embolism on computed tomography pulmonary angiogram scan using a three-dimensional convolutional neural network. European Journal of Radiology. 177, 111586.

[32] Colak, E., Kitamura, F.C., Hobbs, S.B., et al., 2021. The RSNA pulmonary embolism CT dataset. Radiology: Artificial Intelligence. 3(2), e200254.

[33] Ho, J., Jain, A.N., Abbeel, P., 2020. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems. 33, 6840–6851.

[34] Song, Y., Sohl-Dickstein, J., Kingma, D.P., et al., 2020. Score-based generative modeling through stochastic differential equations. arXiv preprint. arXiv:2011.13456.

[35] Vaswani, A., Shazeer, N., Parmar, N., et al., 2017. Attention is all you need. In Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, 4--9 December 2017; pp. 6000–6010.

[36] Ilse, M., Tomczak, J., Welling, M., 2018. Attention-based deep multiple instance learning. Proceedings of the 35th International Conference on Machine Learning. 80, 2127–2136.

[37] Goodfellow, I., Bengio, Y., Courville, A., 2016. Deep Learning. MIT Press: Cambridge, MA, USA.

Downloads

How to Cite

Abdelhamid, A., Moustafa, H. E.-D., B. Nafea, H., H. Abdelhay, E., M. Abo-Zahhad, M., & El-Ghamry, A. (2026). DPHT-Net: A Novel Self-Supervised Hybrid CNN-Transformer Approach for Automated Pulmonary Embolism Classification in CT Pulmonary Angiogram Scans. Journal of Electronic & Information Systems, 8(1), 97–108. https://doi.org/10.30564/jeis.v8i1.13224

Issue

Vol. 8 , Iss. 1 (April 2026): In Progress

Article Type

ARTICLE

License

Copyright © 2026 Abeer Abdelhamid, Hossam El-Din Moustafa, Hala B. Nafea, Ehab H. Abdelhay, Mohammed M. Abo-Zahhad, Amir El-Ghamry

Creative Commons License

This is an open access article under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) License.

Crossref
Scopus
Google Scholar
Europe PMC