Dual-branch hybrid-feature learning for imbalance-aware paediatric colonoscopy decision support with interpretable multi-scale networks

Document Type : Original Article

Author

Department of Biomedical Engineering, Meybod University, Meybod, Iran

10.22091/jemsc.2026.14688.1327

Abstract

Colorectal neoplasia, while rare, is high-risk in children. As such, colonoscopic detection of paediatric polyps is of clinical importance. Current deep learning–based computer-aided systems are trained on adult data and do not provide solutions for class imbalance, cross-centre generalisation or interpretability in paediatrics. We introduce an imbalance-aware dual-branch hybrid-features (IDHF) framework for polyp versus non-polyp classification. IDHF, based on a ResNet-50 backbone, is equipped with complementary texture branch to augment deep semantic features, adaptive gating for late fusion and optimisation of a class-weighted loss function with false-negative penalty and branch-agreement regularisation. The proposed model is trained on CP-CHILD-A (8,000 images, 1,000 polyp/7,000 non-polyp) and tested without fine-tuning on CP-CHILD-B (1,500 images, 400 polyp/1,100 non-polyp) on another platform. ResNet-50+IDHF achieves 99.6% accuracy, 100.0% sensitivity, 99.5% specificity, 96.8% precision and a 98.4% F1-score on CP-CHILD-A. ResNet-50+IDHF achieves 99.5% accuracy, 99.2% sensitivity, 99.6% specificity, 99.0% precision and a 99.1% F1-score on CP-CHILD-B, supporting a robust and interpretable solution for computer-aided paediatric polyp detection.

Keywords

Main Subjects

References

Ahamed, M. F., Islam, M. R., Nahiduzzaman, M., Chowdhury, M. E. H., Alqahtani, A., & Murugappan, M. (2024). Automated colorectal polyps detection from endoscopic images using MultiResUNet framework with attention-guided segmentation. Human-Centric Intelligent Systems, 4, 299–315. https://doi.org/10.1007/s44230-024-00067-1
Altamimi, E., Odeh, Y., Al-Quraan, T., Mohamed, E., & Rawabdeh, N. (2022). Diagnostic and therapeutic outcomes of pediatric colonoscopies in Jordanian children. Journal of Pediatric and Neonatal Individualized Medicine, 11(2), e110211. https://doi.org/10.7363/110211
Bian, H., Jiang, M., & Qian, J. (2023). The investigation of constraints in implementing robust AI colorectal polyp detection for sustainable healthcare system. PLOS ONE, 18(7), e0288376. https://doi.org/10.1371/journal.pone.0288376
Chen, S., Lu, S., Tang, Y., Wang, D., Sun, X., Yi, J., Liu, B., Cao, Y., Chen, Y., & Liu, X. (2022). A machine learning-based system for real-time polyp detection (DeFrame). Frontiers in Medicine, 9, 852553. https://doi.org/10.3389/fmed.2022.852553
Corley, D. A., Jensen, C. D., Marks, A. R., Zhao, W. K., Lee, J. K., Doubeni, C. A., Zauber, A. G., de Boer, J., Fireman, B. H., Schottinger, J. E., Quinn, V. P., Park, C. H., Ghai, N. R., Levin, T. R., & Quesenberry, C. P., Jr. (2014). Adenoma detection rate and risk of colorectal cancer and death. The New England Journal of Medicine, 370(14), 1298–1306. https://doi.org/10.1056/NEJMoa1309086
Dereci, S., Koca, T., & Akçam, M. (2021). To evaluate the diagnostic impact of gastrointestinal symptoms in pediatric patients with inflammatory bowel disease and polyp. Turkish Journal of Pediatric Disease, 15(6), 470–475. https://doi.org/10.12956/tchd.800732
Gelu-Simeon, M., Manole, S., Saurin, J.-C., Ponchon, T., Pujol, B., & Istrate, S. (2025). Deep learning model applied to real-time delineation of colorectal polyps in colonoscopy videos: The RTPoDeMo study. BMC Medical Informatics and Decision Making, 25, 206. https://doi.org/10.1186/s12911-025-03047-y
Hamza, A., Bilal, M., Ramzan, M., & Malik, N. (2025). Effectiveness of encoder–decoder deep learning approach for colorectal polyp segmentation in colonoscopy images. Applied Intelligence, 55, Article 290. https://doi.org/10.1007/s10489-024-06167-6
Handa, P., Goel, N., Indu, S., & Gunjan, D. (2023). Automatic detection of colorectal polyps with mixed convolutions and its occlusion testing. Neural Computing and Applications, 35, 19409–19426. https://doi.org/10.1007/s00521-023-08762-z
Hsu, C.-M., Hsu, C.-C., Hsu, Z.-M., Shih, F.-Y., Chang, M.-L., & Chen, T.-H. (2021). Colorectal polyp image detection and classification through grayscale images and deep learning. Sensors, 21(18), 5995. https://doi.org/10.3390/s21185995
Jha, D., Tomar, N. K., Johansen, H. D., Johansen, D., Rittscher, J., Riegler, M. A., & Halvorsen, P. (2021). Real-time polyp detection, localization and segmentation in colonoscopy using deep learning. IEEE Access, 9, 40496–40510. https://doi.org/10.1109/ACCESS.2021.3063716
Kay, M., Eng, K., & Wyllie, R. (2015). Colonic polyps and polyposis syndromes in pediatric patients. Current Opinion in Pediatrics, 27(5), 634–641. https://doi.org/10.1097/mop.0000000000000265
Krenzer, A., Heil, S., Fitting, D., Matti, S., Zoller, W. G., Hann, A., & Puppe, F. (2023). Automated classification of polyps using deep learning architectures and few-shot learning. BMC Medical Imaging, 23, 59. https://doi.org/10.1186/s12880-023-01007-4
Lee, Y. J., & Park, J. H. (2019). The most common cause of lower gastrointestinal bleeding without other symptoms in children is colonic polyp: Is total colonoscopy needed? Clinical Endoscopy, 52(3), 207–208. https://doi.org/10.5946/ce.2019.084
Li, Q., Yang, G., Chen, Z., Huang, B., Chen, L., Xu, D., & Zhou, X. (2021). Colonoscopy polyp detection and classification: Dataset creation and comparative evaluations. PLOS ONE, 16(8), e0255809. https://doi.org/10.1371/journal.pone.0255809
Maas, M. H. J., Neumann, H., Shirin, H., Katz, L. H., Benson, A. A., Kahloon, A., Soons, E., Hazzan, R., Landsman, M. J., Lebwohl, B., Lewis, S. K., Sivanathan, V., Ngamruengphong, S., Jacob, H., & Siersema, P. D. (2024). A computer-aided polyp detection system in screening and surveillance colonoscopy: An international, multicentre, randomised, tandem trial. The Lancet Digital Health, 6(3), e157–e165. https://doi.org/10.1016/S2589-7500(23)00242-X
Mozaffari, J., Amirkhani, A., & Shokouhi, S. B. (2024). ColonGen: An efficient polyp segmentation system for generalization improvement using a new comprehensive dataset. Physical and Engineering Sciences in Medicine, 47, 309–325. https://doi.org/10.1007/s13246-023-01368-8
National Cancer Institute. (2024, February 11). Childhood colorectal cancer treatment (PDQ®): Health professional version. https://www.cancer.gov/types/colorectal/hp/child-colorectal-treatment-pdq
Nur-A-Alam, M., Uddin, K. M. M., Manu, M. M. R., Rahman, M. M., & Nasir, M. K. (2024). An automatic system to detect colorectal polyp using hybrid fused method from colonoscopy images. Intelligent Systems with Applications, 22, 200342. https://doi.org/10.1016/j.iswa.2024.200342
Pacal, I., & Karaboga, D. (2021). A robust real-time deep learning-based automatic polyp detection system in colonoscopy videos. Computers in Biology and Medicine, 134, 104519. https://doi.org/10.1016/j.compbiomed.2021.104519
Pacal, I., Karaman, A., Karaboga, D., Akay, B., Basturk, A., Nalbantoglu, U., & Coskun, S. (2022). An efficient real-time colonic polyp detection with YOLO algorithms trained by using negative samples and large datasets. Computers in Biology and Medicine, 141, 105031. https://doi.org/10.1016/j.compbiomed.2021.105031
Raseena, T. P., Kumar, J., & Balasundaram, S. R. (2024). DeepCPD: Deep learning with vision transformer for colorectal polyp detection. Multimedia Tools and Applications, 83(32), 78183–78206. https://doi.org/10.1007/s11042-024-18607-z
Raseena, T. P., Kumar, J., & Balasundaram, S. R. (2024). Exploring the effectiveness of deep learning architectures for colorectal polyp detection: Performance analysis and insights. SN Computer Science, 5, 452. https://doi.org/10.1007/s42979-024-02825-1
Saad, A. I., Maghraby, F. A., & Badawy, O. M. (2024). PolyDSS: Computer-aided decision support system for multiclass polyp segmentation and classification using deep learning. Neural Computing and Applications, 36, 5031–5057. https://doi.org/10.1007/s00521-023-09358-3
Selvaraj, J., Sadaf, K., Aslam, S. M., & Umapathy, S. (2025). Multiclassification of colorectal polyps from colonoscopy images using AI for early diagnosis. Diagnostics, 15(10), 1285. https://doi.org/10.3390/diagnostics15101285
Shen, M.-H., Huang, C.-C., Chen, Y.-T., Tsai, Y.-J., Liou, F.-M., Chang, S.-C., & Phan, N. N. (2023). Deep learning empowers endoscopic detection and polyps classification: A multiple-hospital study. Diagnostics, 13(8), 1473. https://doi.org/10.3390/diagnostics13081473
Sultan, I., Rodriguez-Galindo, C., El-Taani, H., Pastore, G., Casanova, M., & Gallino, G. (2010). Distinct features of colorectal cancer in children and adolescents: A population-based study of 159 cases. Cancer, 116(3), 758–765. https://doi.org/10.1002/cncr.24777
Sushama, S., & Menon, V. (2024). Flexible colon polyp detection: A dual mode approach for detection and segmentation of colon polyps with optional inpainting for specular highlight mitigation. SN Computer Science, 5, 641. https://doi.org/10.1007/s42979-024-02932-z
Taghiakbari, M., Mori, Y., & von Renteln, D. (2021). Artificial intelligence-assisted colonoscopy: A review of current state of practice and research. World Journal of Gastroenterology, 27(47), 8103–8122. https://doi.org/10.3748/wjg.v27.i47.8103
Tan, J., Yuan, J., Fu, X., & Bai, Y. (2024). Colonoscopy polyp classification via enhanced scattering wavelet convolutional neural network. PLOS ONE, 19(10), e0302800. https://doi.org/10.1371/journal.pone.0302800
Tudela, Y., Majó, M., de la Fuente, N., Galdran, A., Krenzer, A., Puppe, F., et al., (2024). A complete benchmark for polyp detection, segmentation and classification in colonoscopy images. Frontiers in Oncology, 14, 1417862. https://doi.org/10.3389/fonc.2024.1417862
Wang, W., Tian, J., Zhang, C., Luo, Y., Wang, X., & Li, J. (2020). An improved deep learning approach and its applications on colonic polyp images detection. BMC Medical Imaging, 20(1), 83. https://doi.org/10.1186/s12880-020-00460-4
Wen, G., Yan, J., Chen, X., & Bagal, H. A. (2025). Deep learning with refined single candidate optimizer for early polyp detection. Scientific Reports, 15(1), 40483. https://doi.org/10.1038/s41598-025-24374-0
Wu, C.-T., Chen, C.-A., & Yang, Y.-J. (2015). Characteristics and diagnostic yield of pediatric colonoscopy in Taiwan. Pediatrics & Neonatology, 56(5), 334–338. https://doi.org/10.1016/j.pedneo.2015.01.005
Xi, Y., & Xu, P. (2021). Global colorectal cancer burden in 2020 and projections to 2040. Translational Oncology, 14(10), 101174. https://doi.org/10.1016/j.tranon.2021.101174
Zhu, P.-C., Wan, J.-J., Shao, W., Meng, X.-C., & Chen, B.-L. (2024). Colorectal image analysis for polyp diagnosis. Frontiers in Computational Neuroscience, 18, 1356447. https://doi.org/10.3389/fncom.2024.1356447
Send comment about this article
CAPTCHA Image

Files

Share

How to cite

Statistics