The prevalent overlapping phenomenon of cervical cells in clinical samples constitutes a primary barrier limiting the application of deep learning classification models in cervical cancer screening. This study aims to evaluate the adverse impact of cell overlapping on cell classification performance. Based on the YOLO11l-cls classifier, we fine-tuned the model using segmented non-overlapping single cells. The core approach involves comparing the model’s performance on an ideal single-cell test set and a comparative over-lapping cell test set. Results show that the model exhibited outstanding performance under ideal conditions, achieving an AUC of 0.9959 and a precision of 0.9941. However, its performance declined significantly on the overlapping test set: the AUC dropped to 0.8816, and the precision decreased to 0.7368. These data strongly demonstrate the adverse effect of cell overlapping on cell classification, which leads to deterioration in the model’s classification performance. Therefore, addressing the limitations in feature extraction of over-lapping cells is a critical prerequisite for improving the application value of classification models in complex clinical environments.
References
[1]
Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R.L., Soerjomataram, I., et al. (2024) Global Cancer Statistics 2022: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: ACancerJournalforClinicians, 74, 229-263. https://doi.org/10.3322/caac.21834
[2]
Burmeister, C.A., Khan, S.F., Sch？fer, G., Mbatani, N., Adams, T., Moodley, J., et al. (2022) Cervical Cancer Therapies: Current Challenges and Future Perspectives. TumourVirusResearch, 13, Article ID: 200238. https://doi.org/10.1016/j.tvr.2022.200238
[3]
Bouvard, V., Wentzensen, N., Mackie, A., Berkhof, J., Brotherton, J., Giorgi-Rossi, P., et al. (2022) The IARC Perspective on Cervical Cancer Screening. Obstetrical&GynecologicalSurvey, 77, 154-156. https://doi.org/10.1097/ogx.0000000000001017
[4]
Samikhovna, M.K. (2025) Modern Understanding of the Diagnosis and Prevention of Cervical Cancer. Modern Education and Development, 19, 339-349.
[5]
Staats, P.N., Souers, R.J., Nunez, A.L., Li, Z., Kurtycz, D.F.I., Goodrich, K., et al. (2020) The Differential Diagnosis of Reparative Changes and Malignancy: Performance in the College of American Pathologists Pap Education and Proficiency Testing Programs. ArchivesofPathology&LaboratoryMedicine, 144, 846-852. https://doi.org/10.5858/arpa.2019-0298-cp
[6]
Youneszade, N., Marjani, M. and Pei, C.P. (2023) Deep Learning in Cervical Cancer Diagnosis: Architecture, Opportunities, and Open Research Challenges. IEEEAccess, 11, 6133-6149. https://doi.org/10.1109/access.2023.3235833
[7]
Chen, E., Ting, H., Huang Chuah, J. and Zhao, J. (2024) Segmentation of Overlapping Cells in Cervical Cytology Images: A Survey. IEEEAccess, 12, 114170-114189. https://doi.org/10.1109/access.2024.3445371
[8]
Chen, Y., Huang, P., Lin, K., Lin, H., Wang, L., Cheng, C., et al. (2014) Semi-automatic Segmentation and Classification of Pap Smear Cells. IEEEJournalofBiomedicalandHealthInformatics, 18, 94-108. https://doi.org/10.1109/jbhi.2013.2250984
[9]
Sharma, M., Kumar Singh, S., Agrawal, P. and Madaan, V. (2016) Classification of Clinical Dataset of Cervical Cancer Using KNN. IndianJournalofScienceandTechnology, 9, 1-5. https://doi.org/10.17485/ijst/2016/v9i28/98380
[10]
Kusumawardani., L.A., Rulaningtyas, R. and Winarno, W. (2023) Classification of Cervical Cancer Cells Using the K-Nearest Neighbor (KNN) Method Based on Geometric Feature Extraction. AIP Conference Proceedings, 2858, Article ID: 030003. https://doi.org/10.1063/5.0167165
[11]
Sun, G. (2017) Cervical Cancer Diagnosis Based on Random Forest. InternationalJournalofPerformabilityEngineering, 13, 446-457. https://doi.org/10.23940/ijpe.17.04.p12.446457
[12]
Xu, L., Cai, F., Fu, Y. and Liu, Q. (2023) Cervical Cell Classification with Deep-Learning Algorithms. Medical&BiologicalEngineering&Computing, 61, 821-833. https://doi.org/10.1007/s11517-022-02745-3
[13]
Rahaman, M.M., Li, C., Yao, Y., Kulwa, F., Wu, X., Li, X., et al. (2021) DeepCervix: A Deep Learning-Based Framework for the Classification of Cervical Cells Using Hybrid Deep Feature Fusion Techniques. ComputersinBiologyandMedicine, 136, Article ID: 104649. https://doi.org/10.1016/j.compbiomed.2021.104649
[14]
Chen, H., Liu, J., Wen, Q., Zuo, Z., Liu, J., Feng, J., et al. (2021) CytoBrain: Cervical Cancer Screening System Based on Deep Learning Technology. JournalofComputerScienceandTechnology, 36, 347-360. https://doi.org/10.1007/s11390-021-0849-3
[15]
Maurya, R., Nath Pandey, N. and Kishore Dutta, M. (2023) VisionCervix: Papanicolaou Cervical Smears Classification Using Novel CNN-Vision Ensemble Approach. BiomedicalSignalProcessingandControl, 79, Article ID: 104156. https://doi.org/10.1016/j.bspc.2022.104156
[16]
Zhang, Y., Ning, C. and Yang, W. (2025) An Automatic Cervical Cell Classification Model Based on Improved DenseNet121. ScientificReports, 15, Article No. 3240. https://doi.org/10.1038/s41598-025-87953-1
[17]
Deepa, T.P. and Rao, A.N. (2024) Classification of Normal and Abnormal Overlapped Squamous Cells in Pap Smear Image. InternationalJournalofSystemAssuranceEngineeringandManagement, 15, 519-531. https://doi.org/10.1007/s13198-022-01805-z
[18]
Mulmule, P.V. and Kanphade, R.D. (2021) Classification of Overlapping Cells in Microscopic Cervical Images: A Transfer Learning Approach. 2021 Asian Conference on Innovation in Technology (ASIANCON), Pune, 27-29 August 2021, 1-7. https://doi.org/10.1109/asiancon51346.2021.9544587
[19]
Nguyen, D.H. and Nguyen, T.N.L. (2025) A Study on Semi-Supervised Solutions for Overlapped Cell Classification. Vietnam-Korea University of Information and Communication Technology, Working Paper. https://elib.vku.udn.vn/handle/123456789/5009
[20]
Mulmule, P.V. and Kanphade, R.D. (2022) Classification of Cervical Cytology Overlapping Cell Images with Transfer Learning Architectures. BiomedicalandPharmacologyJournal, 15, 277-284. https://doi.org/10.13005/bpj/2364
[21]
Yu, S., Feng, X., Wang, B., Dun, H., Zhang, S., Zhang, R., et al. (2021) Automatic Classification of Cervical Cells Using Deep Learning Method. IEEEAccess, 9, 32559-32568. https://doi.org/10.1109/access.2021.3060447
[22]
Pachitariu, M., Rariden, M. and Stringer, C. (2025) Cellpose-SAM: Superhuman Generalization for Cellular Segmentation. bioRxiv. https://doi.org/10.1101/2025.04.28.651001
[23]
Stringer, C., Wang, T., Michaelos, M. and Pachitariu, M. (2021) Cellpose: A Generalist Algorithm for Cellular Segmentation. NatureMethods, 18, 100-106. https://doi.org/10.1038/s41592-020-01018-x
[24]
Khanam, R. and Hussain, M. (2024) YOLOv11: An Overview of the Key Architectural Enhancements. arXiv: 2410.17725.
[25]
Redmon, J., Divvala, S., Girshick, R. and Farhadi, A. (2016) You Only Look Once: Unified, Real-Time Object Detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 779-788. https://doi.org/10.1109/cvpr.2016.91
[26]
Jegham, N., Koh, C.Y., Abdelatti, M. and Hendawi, A. (2024) YOLO Evolution: A Comprehensive Benchmark and Architectural Review of YOLOv12, YOLO11, and Their Previous Versions. arXiv: 2411.00201v3.
[27]
Rasheed, A.F. and Zarkoosh, M. (2025) YOLOv11 Optimization for Efficient Resource Utilization. TheJournalofSupercomputing, 81, Article No. 1085. https://doi.org/10.1007/s11227-025-07520-3
