{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T12:46:23Z","timestamp":1785415583078,"version":"3.56.0"},"reference-count":44,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T00:00:00Z","timestamp":1776643200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004410","name":"Scientific and Technological Research Council of Turkey","doi-asserted-by":"crossref","award":["122E164"],"award-info":[{"award-number":["122E164"]}],"id":[{"id":"10.13039\/501100004410","id-type":"DOI","asserted-by":"crossref"}]},{"award":["122E164"],"award-info":[{"award-number":["122E164"]}],"id":[{"id":"https:\/\/ror.org\/04w9kkr77","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005322","name":"Recep Tayyip Erdo\u011fan University Development Foundation","doi-asserted-by":"publisher","award":["020250120150852"],"award-info":[{"award-number":["020250120150852"]}],"id":[{"id":"10.13039\/501100005322","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: The manual analysis of sperm morphology, crucial for male infertility diagnosis, is subjective and time-consuming. Automated methods using deep learning, offer a promising alternative; however, standard deep models are prone to overfitting when applied to small, heavily unbalanced clinical datasets, limiting their generalization capability. This study proposes a knowledge distillation approach that functions as a strong regularizer, improving the robustness of automated sperm morphology analysis. Methods: We utilize soft distillation to transfer knowledge from a set of high-capacity teacher models to a smaller student model (SwinV2-base). The teacher architectures include SwinV2-large, EfficientNetV2-m, and ConvNeXtV2-large. To maximize performance, we investigated two distillation strategies: a single-teacher approach, where the student learns from one specific architecture, and a multi-teacher approach, where the student learns from an averaged response of multiple teachers. The models were trained on the imbalanced Hi-LabSpermMorpho dataset, which comprises 18 different sperm morphology categories derived from three differently stained (BesLab, Histoplus, GBL) sample sets. We adopted a cross-dataset training approach in which the teacher models were fine-tuned using the combination of two stained datasets, and the student model was trained on the third, distinct stained dataset. The global loss function combined cross-entropy loss with Kullback\u2013Leibler divergence, employing the teacher\u2019s soft probabilities to prevent the student from over-confidence. Results: The experimental results demonstrate that the student model trained in a multi-teacher setup with augmentation and soft distillation attains higher accuracies (70.94% on BesLab, 73.61% on Histoplus, 71.63% on GBL) than the baseline models. Conclusions: This approach mitigates challenges associated with data scarcity and heavily unbalanced sperm morphology datasets, providing consistent improvements and offering a highly generalizable solution for clinical diagnostics.<\/jats:p>","DOI":"10.3390\/diagnostics16081230","type":"journal-article","created":{"date-parts":[[2026,4,20]],"date-time":"2026-04-20T13:22:47Z","timestamp":1776691367000},"page":"1230","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Multi-Teacher Knowledge Distillation Framework for Enhancing the Robustness of Automated Sperm Morphology Assessment"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-3694-3696","authenticated-orcid":false,"given":"Osman Emre","family":"Tutay","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, Yildiz Technical University, Istanbul 34220, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1753-2703","authenticated-orcid":false,"given":"Hamza Osman","family":"Ilhan","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, Yildiz Technical University, Istanbul 34220, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5189-3166","authenticated-orcid":false,"given":"Hakk\u0131","family":"Uzun","sequence":"additional","affiliation":[{"name":"Department of Urology, Faculty of Medicine, Recep Tayyip Erdo\u011fan University, Rize 53100, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8988-4321","authenticated-orcid":false,"given":"Merve","family":"Huner Yigit","sequence":"additional","affiliation":[{"name":"Department of Biochemistry, Faculty of Medicine, Recep Tayyip Erdo\u011fan University, Rize 53100, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4591-7368","authenticated-orcid":false,"given":"Gorkem","family":"Serbes","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering, Yildiz Technical University, Istanbul 34220, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Shahzad, S., Ilyas, M., Lali, M.I.U., Rauf, H.T., Kadry, S., and Nasr, E.A. (2023). Sperm abnormality detection using sequential deep neural network. Mathematics, 11.","DOI":"10.3390\/math11030515"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Guo, Y., Li, J., Hong, K., Wang, B., Zhu, W., Li, Y., Lv, T., and Wang, L. (2024). Automated Deep Learning Model for Sperm Head Segmentation, Pose Correction, and Classification. Appl. Sci., 14.","DOI":"10.3390\/app142311303"},{"key":"ref_3","first-page":"283","article-title":"A new method for sperm detection in human semen: Combination of hypothesis testing and local mapping of wavelet sub-bands","volume":"9","author":"Shojaedini","year":"2012","journal-title":"Iran. J. Med. Phys."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"384","DOI":"10.4236\/jbise.2012.57049","article-title":"Fully automatic identification and discrimination of sperm\u2019s parts in microscopic images of stained human semen smear","volume":"5","author":"Bijar","year":"2012","journal-title":"J. Biomed. Sci. Eng."},{"key":"ref_5","unstructured":"Shi, Y., Wang, Y.K., Tian, X.P., Zhang, T.Y., Yao, B., Wang, H., Shao, Y., Wang, C.C., and Zeng, R. (March, January 25). SPEHEATAL: A Cluster-Enhanced Segmentation Method for Sperm Morphology Analysis. Proceedings of the AAAI Conference on Artificial Intelligence, Philadelphia, PA, USA."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Lei, P., Saadat, M., Hassani, M.G., and Shu, C. (2025). Deep Learning Models for Multi-Part Morphological Segmentation and Evaluation of Live Unstained Human Sperm. Sensors, 25.","DOI":"10.3390\/s25103093"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"687607","DOI":"10.1155\/2013\/687607","article-title":"Computer-assisted system with multiple feature fused support vector machine for sperm morphology diagnosis","volume":"2013","author":"Tseng","year":"2013","journal-title":"BioMed Res. Int."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Keskenler, M.F., Ha\u015filo\u011flu, A., \u00d6zyer, G.T., \u00d6zyer, B., and \u015e\u0131m\u015fek, E. (2019). Sperm Detection and Analysis Using Feature Description Algorithms. Proceedings of the 2019 27th Signal Processing and Communications Applications Conference (SIU), IEEE.","DOI":"10.1109\/SIU.2019.8806287"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"57","DOI":"10.33005\/itij.v2i2.36","article-title":"Detection of Abnormal Human Sperm Morphology Using Support Vector Machine (SVM) Classification","volume":"2","author":"Diyasa","year":"2024","journal-title":"Inf. Technol. Int. J."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"893","DOI":"10.1002\/cyto.a.23189","article-title":"Automated analysis of individual sperm cells using stain-free interferometric phase microscopy and machine learning","volume":"91","author":"Mirsky","year":"2017","journal-title":"Cytom. Part A"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"K\u0131l\u0131\u00e7, \u015e. (2025). Deep feature engineering for accurate sperm morphology classification using CBAM-enhanced ResNet50. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0330914"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Liang, B., and Wang, M. (2025). Deep learning-based approach for sperm morphology analysis. BMC Urol., 25.","DOI":"10.1186\/s12894-025-01946-w"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"16360","DOI":"10.3934\/math.2023838","article-title":"A review of different deep learning techniques for sperm fertility prediction","volume":"8","author":"Suleman","year":"2023","journal-title":"AIMS Math."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Zhang, J., Zha, X., Zhou, Y., Cao, Y., and Chen, D. (2022). Improving human sperm head morphology classification with unsupervised anatomical feature distillation. Proceedings of the 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI), IEEE.","DOI":"10.1109\/ISBI52829.2022.9761633"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Iqbal, I., Mustafa, G., and Ma, J. (2020). Deep learning-based morphological classification of human sperm heads. Diagnostics, 10.","DOI":"10.3390\/diagnostics10050325"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1789","DOI":"10.1007\/s11263-021-01453-z","article-title":"Knowledge distillation: A survey","volume":"129","author":"Gou","year":"2021","journal-title":"Int. J. Comput. Vis."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"103177","DOI":"10.1016\/j.bspc.2021.103177","article-title":"Lightweight convolutional neural network with knowledge distillation for cervical cells classification","volume":"71","author":"Chen","year":"2022","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Belinga, A.G., Tekouabou Koumetio, C.S., El Haziti, M., and El Hassouni, M. (2024). Knowledge Distillation in Image Classification: The Impact of Datasets. Computers, 13.","DOI":"10.3390\/computers13080184"},{"key":"ref_19","unstructured":"Hinton, G., Vinyals, O., and Dean, J. (2015). Distilling the knowledge in a neural network. arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009). Imagenet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref_21","unstructured":"Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Zitnick, C.L. (2014). Microsoft coco: Common objects in context. Proceedings of the Computer Vision\u2013ECCV 2014: 13th European Conference, Zurich, Switzerland, 6\u201312 September 2014, Springer. Proceedings, Part V 13."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Sun, C., Shrivastava, A., Singh, S., and Gupta, A. (2017). Revisiting unreasonable effectiveness of data in deep learning era. Proceedings of the IEEE International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV.2017.97"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"196070","DOI":"10.1109\/ACCESS.2024.3521643","article-title":"Hi-LabSpermMorpho: A Novel Expert-Labeled Dataset with Extensive Abnormality Classes for Deep Learning-Based Sperm Morphology Analysis","volume":"12","author":"Aktas","year":"2024","journal-title":"IEEE Access"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"\u015eavkay, O.L., Cesur, E., Yal\u00e7\u0131n, M.E., and Tav\u015fano\u011flu, V. (2014). Sperm Morphology Analysis with CNN based algorithms. Proceedings of the 2014 14th International Workshop on Cellular Nanoscale Networks and Their Applications (CNNA), IEEE.","DOI":"10.1109\/CNNA.2014.6888647"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Alqu\u00e9zar-Baeta, C., Gimeno-Martos, S., Miguel-Jim\u00e9nez, S., Santolaria, P., Y\u00e1niz, J., Palac\u00edn, I., Casao, A., Cebri\u00e1n-P\u00e9rez, J.\u00c1., Mui\u00f1o-Blanco, T., and P\u00e9rez-P\u00e9, R. (2019). OpenCASA: A new open-source and scalable tool for sperm quality analysis. PLoS Comput. Biol., 15.","DOI":"10.1371\/journal.pcbi.1006691"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"202","DOI":"10.1016\/S0015-0282(16)55640-4","article-title":"A new computerized method of reading sperm morphology (strict criteria) is as efficient as technician reading","volume":"59","author":"Kruger","year":"1993","journal-title":"Fertil. Steril."},{"key":"ref_27","unstructured":"Yi, W.J., Park, K.S., and Paick, J.S. (1998). Morphological classification of sperm heads using artificial neural networks. MEDINFO\u201998, IOS Press."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"27345","DOI":"10.1007\/s11042-024-20188-w","article-title":"Advancements in automated sperm morphology analysis: A deep learning approach with comprehensive classification and model evaluation","volume":"84","author":"Maalej","year":"2025","journal-title":"Multimed. Tools Appl."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"101459","DOI":"10.1016\/j.imu.2024.101459","article-title":"Deep learning-based sperm motility and morphology estimation on stacked color-coded MotionFlow","volume":"45","author":"Adinugroho","year":"2024","journal-title":"Inform. Med. Unlocked"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Nguyen, V.D., Ngo, T.T.H., and Duong, L.M. (2024). Automated Sperm Morphology Analysis with Deep Learning and Feature Extraction. Proceedings of the 2024 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES), IEEE.","DOI":"10.1109\/IECBES61011.2024.10990869"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1007\/s11517-025-03418-7","article-title":"Deep learning-based morphological analysis of human sperm","volume":"64","author":"Xu","year":"2026","journal-title":"Med. Biol. Eng. Comput."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"2583010","DOI":"10.1080\/20565623.2025.2583010","article-title":"Deep-learning based model for sperm morphology assessment using the SMD\/MSS dataset","volume":"11","author":"Abdelkefi","year":"2025","journal-title":"Future Sci. OA"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1016\/j.compbiomed.2017.10.009","article-title":"A dictionary learning approach for human sperm heads classification","volume":"91","author":"Shaker","year":"2017","journal-title":"Comput. Biol. Med."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1016\/j.compbiomed.2017.03.004","article-title":"Gold-standard for computer-assisted morphological sperm analysis","volume":"83","author":"Chang","year":"2017","journal-title":"Comput. Biol. Med."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"182","DOI":"10.1016\/j.compbiomed.2019.04.030","article-title":"A novel deep learning method for automatic assessment of human sperm images","volume":"109","author":"Javadi","year":"2019","journal-title":"Comput. Biol. Med."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Turkoglu, A.K., Serbes, G., Uzun, H., Aktas, A., Yigit, M.H., and Ilhan, H.O. (2025). Category-aware two-stage divide-and-ensemble framework for sperm morphology classification. Diagnostics, 15.","DOI":"10.3390\/diagnostics15172234"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"3048","DOI":"10.1109\/TPAMI.2021.3055564","article-title":"Knowledge distillation and student-teacher learning for visual intelligence: A review and new outlooks","volume":"44","author":"Wang","year":"2021","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"2347978","DOI":"10.1080\/21681163.2024.2347978","article-title":"Less-supervised learning with knowledge distillation for sperm morphology analysis","volume":"12","author":"Nabipour","year":"2024","journal-title":"Comput. Methods Biomech. Biomed. Eng. Imaging Vis."},{"key":"ref_39","unstructured":"World Health Organization (2021). WHO Laboratory Manual for the Examination and Processing of Human Semen, World Health Organization."},{"key":"ref_40","unstructured":"TorchVision Maintainers and Contributors (2026, April 16). TorchVision: PyTorch\u2019s Computer Vision Library, 2016. Available online: https:\/\/github.com\/pytorch\/vision."},{"key":"ref_41","unstructured":"Tan, M., and Le, Q. (2021, January 18\u201324). Efficientnetv2: Smaller models and faster training. Proceedings of the International Conference on Machine Learning, Virtual."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Woo, S., Debnath, S., Hu, R., Chen, X., Liu, Z., Kweon, I.S., and Xie, S. (2023, January 17\u201324). Convnext v2: Co-designing and scaling convnets with masked autoencoders. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.","DOI":"10.1109\/CVPR52729.2023.01548"},{"key":"ref_43","first-page":"108","article-title":"A Comprehensive Review of ConvNeXt Architecture in Image Classification: Performance, Applications, and Prospects","volume":"2","author":"Ilham","year":"2026","journal-title":"IJACI Int. J. Adv. Comput. Inform."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Hosseini, A., and Serag, A. (2025). Is synthetic data generation effective in maintaining clinical biomarkers? Investigating diffusion models across diverse imaging modalities. Front. Artif. Intell., 7.","DOI":"10.3389\/frai.2024.1454441"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/8\/1230\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T04:15:15Z","timestamp":1776831315000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/8\/1230"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,20]]},"references-count":44,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["diagnostics16081230"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16081230","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,20]]}}}