{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T04:05:17Z","timestamp":1784261117956,"version":"3.55.0"},"reference-count":37,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T00:00:00Z","timestamp":1784160000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computers"],"abstract":"<jats:p>Cervical cancer remains a major global public health challenge, and the accurate classification of cervical transformation zones (TZs) constitutes a critical step in early detection and clinical decision-making. However, distinguishing between Type 2 and Type 3 transformation zones remains particularly challenging due to their high morphological similarity and the inherent interobserver variability associated with colposcopic assessment. In this study, we propose a novel Dual-Track Specialist Feature Fusion and Meta-Learning Stacking Ensemble architecture for the automated classification of cervical transformation zones using the Intel &amp; MobileODT Cervical Cancer Screening dataset. The proposed framework integrates a global feature extractor based on ResNet50 (Gatekeeper) with a visual specialist based on InceptionResNetV2, trained exclusively on the most diagnostically ambiguous cases (Type 2 and Type 3). The extracted features are fused and processed through a multi-level stacking scheme composed of Multilayer Perceptron (MLP), Support Vector Machine (SVM), Gradient Boosting (GB), XGBoost, and LightGBM classifiers at the base level, followed by an XGBoost meta-learner and a clinically guided probability calibration strategy designed to maximize diagnostic sensitivity. Experimental results demonstrate a peak overall accuracy of 91.22%, substantially outperforming the baseline ResNet50 model (70%). Furthermore, the proposed system achieved Recall values of 0.90, 0.90, and 0.94 for Type 1, Type 2, and Type 3 transformation zones, respectively, highlighting its ability to accurately identify diagnostically challenging cases. Ablation studies, Grad-CAM visualizations, and external-image validation experiments confirm that the proposed architecture improves discrimination between ambiguous categories, learns clinically meaningful representations, and maintains strong generalization capability across heterogeneous scenarios. These findings demonstrate the potential of visual specialization and calibrated meta-learning strategies for the development of artificial intelligence-assisted colposcopic decision-support systems.<\/jats:p>","DOI":"10.3390\/computers15070450","type":"journal-article","created":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T03:12:38Z","timestamp":1784257958000},"page":"450","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Dual-Track Specialist Feature Fusion and Meta-Learning Stacking Ensemble for Cervical Transformation Zone Classification in Colposcopy"],"prefix":"10.3390","volume":"15","author":[{"given":"Edgar Fabi\u00e1n","family":"Rivera-Guzm\u00e1n","sequence":"first","affiliation":[{"name":"GI-IATa, UNESCO Chair on Support Technologies for Educational Inclusion, Universidad Polit\u00e9cnica Salesiana, Cuenca 010103, Ecuador"},{"name":"Carrera de Agroindustrias, Universidad Estatal de Bol\u00edvar, Guaranda 020150, Ecuador"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7645-8793","authenticated-orcid":false,"given":"Vladimir Espartaco","family":"Robles-Bykbaev","sequence":"additional","affiliation":[{"name":"GI-IATa, UNESCO Chair on Support Technologies for Educational Inclusion, Universidad Polit\u00e9cnica Salesiana, Cuenca 010103, Ecuador"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2545-4733","authenticated-orcid":false,"given":"Bernardo J.","family":"Vega-Crespo","sequence":"additional","affiliation":[{"name":"Facultad de Ciencias M\u00e9dicas, Universidad de Cuenca, Cuenca 010203, Ecuador"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3708-6501","authenticated-orcid":false,"given":"Veronique","family":"Verhoeven","sequence":"additional","affiliation":[{"name":"Family Medicine and Population Health, University of Antwerp, 2610 Antwerp, Belgium"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,16]]},"reference":[{"key":"ref_1","first-page":"e1756","article-title":"Global, regional and national burden, incidence, and mortality of cervical cancer","volume":"6","author":"Momenimovahed","year":"2023","journal-title":"Cancer Rep."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Azhari, H.A., Chawdhury, M., Islam, F., Zakaria, G.A., Dalal, K., and Reza, H.M. (2025). Visual inspection with acetic acid and colposcopy: Screening of cervical cancer in resource-limited healthcare settings. Front. Glob. Women\u2019s Health, 6.","DOI":"10.3389\/fgwh.2025.1684205"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"475","DOI":"10.1097\/PGP.0000000000000277","article-title":"Endocervicosis of the Uterine Cervix","volume":"35","author":"Mobarki","year":"2016","journal-title":"Int. J. Gynecol. Pathol."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Gireesh, N., Nair, R.P., Nemmini, S., Vani, R., and Panicker, J.V. (2024, January 21\u201323). Automated System for Efficient Cervical Cancer Screening Using Cervicogram Images. Proceedings of the 2024 International Conference on Intelligent Computing and Sustainable Innovations in Technology (IC-SIT), Bhubaneswar, India.","DOI":"10.1109\/IC-SIT63503.2024.10862079"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"955","DOI":"10.1080\/17434440.2024.2407549","article-title":"Digital colposcopy image analysis techniques requirements and their role in clinical diagnosis: A systematic review","volume":"21","author":"Tamang","year":"2024","journal-title":"Expert Rev. Med. Devices"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Alzubaidi, L., Fadhel, M.A., Al-Shamma, O., Zhang, J., Santamar\u00eda, J., Duan, Y., and Oleiwi, S.R. (2020). Towards a better understanding of transfer learning for medical imaging: A case study. Appl. Sci., 10.","DOI":"10.3390\/app10134523"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Hell\u00edn, C.J., Olmedo, A.A., Valledor, A., G\u00f3mez, J., L\u00f3pez-Ben\u00edtez, M., and Tayebi, A. (2024). Unraveling the Impact of Class Imbalance on Deep-Learning Models for Medical Image Classification. Appl. Sci., 14.","DOI":"10.3390\/app14083419"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Rivera-Guzm\u00e1n, E.F., Guerrero-V\u00e1squez, L.F., and Robles-Bykbaev, V.E. (2026). Quantization of Deep Neural Networks for Medical Image Analysis: A Systematic Review and Meta-Analysis. Technologies, 14.","DOI":"10.3390\/technologies14010076"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Guo, P., Xue, Z., Mtema, Z., Yeates, K., Ginsburg, O., Demarco, M., Long, L.R., Schiffman, M., and Antani, S. (2020). Ensemble deep learning for cervix image selection toward improving reliability in automated cervical precancer screening. Diagnostics, 10.","DOI":"10.3390\/diagnostics10070451"},{"key":"ref_10","first-page":"41","article-title":"Development of image preprocessing toolbox to boost cervical cancer detection","volume":"Volume 13307","author":"Ekem","year":"2025","journal-title":"Optics and Biophotonics in Low-Resource Settings XI"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Guo, P., Xue, Z., Long, L.R., and Antani, S. (2020). Cross-dataset evaluation of deep learning networks for uterine cervix segmentation. Diagnostics, 10.","DOI":"10.3390\/diagnostics10010044"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Gaona, Y.J., Malla, D.C., Crespo, B.V., Vicu\u00f1a, M.J., Neira, V.A., D\u00e1vila, S., and Verhoeven, V. (2022). Radiomics Diagnostic Tool Based on Deep Learning for Colposcopy Image Classification. Diagnostics, 12.","DOI":"10.3390\/diagnostics12071694"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Hemalatha, K., and Vetriselvi, V. (2022). A Survey on Cervical Cancer Detection and Classification Using Deep Learning. IFIP Advances in Information and Communication Technology, Springer International Publishing.","DOI":"10.1007\/978-3-031-16364-7_2"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e5","DOI":"10.4108\/eai.12-4-2021.169183","article-title":"Cervix image classification for prognosis of cervical cancer using deep neural network with transfer learning","volume":"7","author":"Dhawan","year":"2021","journal-title":"EAI Endorsed Trans. Pervasive Health Technol."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Akbar, H., Anwar, N., Rohajawati, S., Yulfitri, A., and Kaurani, H.S. (2021). Optimizing AlexNet using swarm intelligence for cervical cancer classification. 2021 International Symposium on Electronics and Smart Devices: Intelligent Systems for Present and Future Challenges, ISESD 2021, IEEE.","DOI":"10.1109\/ISESD53023.2021.9501561"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1007\/s11063-025-11770-w","article-title":"Cervical Cancer Classification Using Deep Learning Approach Using Colposcopy Images","volume":"57","author":"Dayalane","year":"2025","journal-title":"Neural Process. Lett."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Darwish, M., Altabel, M.Z., and Abiyev, R.H. (2023). Enhancing Cervical Pre-Cancerous Classification Using Advanced Vision Transformer. Diagnostics, 13.","DOI":"10.3390\/diagnostics13182884"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Neeli, J., Patil, S.S., Kapadiya, B., and Patil, S. (2025). Deep Learning for Cervical Cell Analysis Toward Automatic Diagnosis. Artificial Intelligence: Theory and Applications, Springer Nature.","DOI":"10.1007\/978-981-96-1918-4_14"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"163","DOI":"10.2147\/MDER.S366303","article-title":"Cervix Type and Cervical Cancer Classification System Using Deep Learning Techniques","volume":"15","author":"Habtemariam","year":"2022","journal-title":"Med. Devices Evid. Res."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"AlMohimeed, A., Saleh, H., Mostafa, S., Saad, R.M.A., and Talaat, A.S. (2023). Cervical Cancer Diagnosis Using Stacked Ensemble Model and Optimized Feature Selection: An Explainable Artificial Intelligence Approach. Computers, 12.","DOI":"10.3390\/computers12100200"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Hong, Z., Xiong, J., Yang, H., and Mo, Y.K. (2024). Lightweight Low-Rank Adaptation Vision Transformer Framework for Cervical Cancer Detection and Cervix Type Classification. Bioengineering, 11.","DOI":"10.3390\/bioengineering11050468"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"732","DOI":"10.1109\/JBHI.2025.3601059","article-title":"Dual-MFNet: AI-Driven Dual-Scale Multimodal Fusion With State Space Networks for Personalized MRI Synthesis","volume":"30","author":"Lyu","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"20552076251327945","DOI":"10.1177\/20552076251327945","article-title":"Interpretable artificial intelligence (AI) for cervical cancer risk analysis leveraging stacking ensemble and expert knowledge","volume":"11","author":"Roy","year":"2025","journal-title":"Digit. Health"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Muksimova, S., Umirzakova, S., Shoraimov, K., Baltayev, J., and Cho, Y.I. (2024). Novelty Classification Model Use in Reinforcement Learning for Cervical Cancer. Cancers, 16.","DOI":"10.3390\/cancers16223782"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Susan, M.B.J., and Subashini, P. (2023). Detection of Specular Reflection on the Smart Colposcopy Images Using Fine-Tuned U-Net Model. Engineering, Science, and Sustainability, CRC Press.","DOI":"10.4324\/9781003388982-4"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1186\/s40537-019-0192-5","article-title":"Survey on deep learning with class imbalance","volume":"6","author":"Johnson","year":"2019","journal-title":"J. Big Data"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A. (2017, January 4\u20139). Inception-v4, inception-ResNet and the impact of residual connections on learning. Proceedings of the 31st AAAI Conference on Artificial Intelligence, AAAI 2017, San Francisco, CA, USA.","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1109\/MSP.2017.2738401","article-title":"Deep multimodal learning: A survey on recent advances and trends","volume":"34","author":"Ramachandram","year":"2017","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_29","unstructured":"Kohavi, R. (1995, January 20\u201325). A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection. Proceedings of the IJCAI International Joint Conference on Artificial Intelligence, Montreal, QC, Canada."},{"key":"ref_30","unstructured":"Goodfellow, I. (2016). Deep Learning-Ian Goodfellow, Yoshua Bengio, Aaron Courville-Google Books, MIT Press."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support-Vector Networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Chen, T., and Guestrin, C. (2016, January 13\u201317). XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939785"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1263","DOI":"10.1109\/TKDE.2008.239","article-title":"Learning from imbalanced data","volume":"21","author":"He","year":"2009","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"271","DOI":"10.1613\/jair.594","article-title":"Issues in stacked generalization","volume":"10","author":"Ting","year":"1999","journal-title":"J. Artif. Intell. Res."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1109\/MCAS.2006.1688199","article-title":"Ensemble based systems in decision making","volume":"6","author":"Polikar","year":"2006","journal-title":"IEEE Circuits Syst. Mag."},{"key":"ref_36","unstructured":"Weatherwax, J.L., and Epstein, D. (2026, April 09). A Solution Manual and Notes for: The Elements of Statistical Learning by Jerome Friedman, Trevor Hastie, and Robert Tibshirani. Available online: https:\/\/waxworksmath.com\/Authors\/G_M\/Hastie\/WriteUp\/Weatherwax_Epstein_Hastie_Solution_Manual.pdf."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Van Calster, B., McLernon, D.J., van Smeden, M., Wynants, L., and Steyerberg, E.W. (2019). Calibration: The Achilles heel of predictive analytics. BMC Med., 17.","DOI":"10.1186\/s12916-019-1466-7"}],"container-title":["Computers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-431X\/15\/7\/450\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T03:40:30Z","timestamp":1784259630000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-431X\/15\/7\/450"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,16]]},"references-count":37,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["computers15070450"],"URL":"https:\/\/doi.org\/10.3390\/computers15070450","relation":{},"ISSN":["2073-431X"],"issn-type":[{"value":"2073-431X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,16]]}}}