{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T12:14:40Z","timestamp":1780402480110,"version":"3.54.1"},"reference-count":32,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Biosensors"],"abstract":"<jats:p>Herbal medicines represent a significant global market, yet food safety remains threatened by counterfeit products morphologically resembling authentic samples. Models trained on limited datasets are prone to shortcut learning, relying on superficial features rather than intrinsic morphological characteristics. This study identified size-based shortcut learning as a critical factor degrading the classification of Ziziphus jujuba Mill. var. spinosa and its counterfeit Ziziphus mauritiana Lam., and demonstrated that focal loss alone can effectively mitigate this issue. Models trained on the internal dataset were evaluated on an external dataset acquired with the Herb-X. On the internal test set, all configurations achieved high classification accuracies (\u226598%), thereby obscuring meaningful differences in external generalization. However, consistent performance degradation was observed on the external dataset. The cross-entropy model trained on background-removed data dropped to 82.08 \u00b1 10.97%, while size-normalized models recovered to 84.17 \u00b1 10.15% (upsizing) and 88.94 \u00b1 6.76% (downsizing), confirming that suppressing size shortcuts improves external generalization. The focal loss model, without any preprocessing, achieved 90.88 \u00b1 2.71%, reducing the internal\u2013external generalization gap from 16.18 to 8.11 percentage points. Grad-CAM++ and loss analyses confirmed that the focal loss model attended to intrinsic morphological features rather than object size. This study provides a practical, preprocessing-free approach for reliable herbal-medicine authentication in field conditions.<\/jats:p>","DOI":"10.3390\/bios16060320","type":"journal-article","created":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T11:25:31Z","timestamp":1780399531000},"page":"320","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["CNN-Based Classification of Ziziphus Seeds with Focal Loss for Overcoming Size-Based Shortcut Learning"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-6470-1568","authenticated-orcid":false,"given":"Yea-Jin","family":"Park","sequence":"first","affiliation":[{"name":"Department of Smart Farm Science, Kyung Hee University, Yongin 17104, Republic of Korea"},{"name":"Interdisciplinary Program in IT-Bio Convergence System, Kyung Hee University, Yongin 17104, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9826-2695","authenticated-orcid":false,"given":"Dae-Hyun","family":"Jung","sequence":"additional","affiliation":[{"name":"Department of Smart Farm Science, Kyung Hee University, Yongin 17104, Republic of Korea"},{"name":"Interdisciplinary Program in IT-Bio Convergence System, Kyung Hee University, Yongin 17104, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,2]]},"reference":[{"key":"ref_1","first-page":"78","article-title":"Comparative evaluation of total phenols, flavonoids content and antioxidant potential of leaf and fruit extracts of Omani Ziziphus jujuba L.","volume":"18","author":"Hossain","year":"2016","journal-title":"Pac. Sci. Rev. A Nat. Sci. Eng."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"35","DOI":"10.14218\/FIM.2023.00086","article-title":"Exploring the safety, efficacy, and bioactivity of herbal medicines: Bridging traditional wisdom and modern science in healthcare","volume":"3","author":"Balkrishna","year":"2024","journal-title":"Future Integr. Med."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1016\/j.copbio.2013.09.010","article-title":"DNA barcoding of medicinal plant material for identification","volume":"25","author":"Techen","year":"2014","journal-title":"Curr. Opin. Biotechnol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"651","DOI":"10.1007\/s40264-017-0527-0","article-title":"Species adulteration in the herbal trade: Causes, consequences and mitigation","volume":"40","author":"Srirama","year":"2017","journal-title":"Drug Saf."},{"key":"ref_5","unstructured":"Seo, K.-W. (2022). The Dispensatory on the Visual and Organoleptic Examination of Herbal Medicine, National Institute of Food and Drug Safety Evaluation."},{"key":"ref_6","first-page":"72","article-title":"Effects of the Zizyphus jujuba seed extract on the lipid components in hyperlipidemic rats","volume":"7","author":"Kim","year":"2002","journal-title":"J. Food Sci. Nutr."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"51","DOI":"10.2147\/JEP.S379805","article-title":"Phytochemical screening and antimicrobial activity evaluation of selected medicinal plants in Ethiopia","volume":"15","author":"Dubale","year":"2023","journal-title":"J. Exp. Pharmacol."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"114013","DOI":"10.1016\/j.postharvbio.2025.114013","article-title":"Effect of low-temperature storage regimes on the UV-B sensitivity of apple anthocyanin","volume":"232","author":"Duan","year":"2026","journal-title":"Postharvest Biol. Technol."},{"key":"ref_9","first-page":"100179","article-title":"Artificial intelligence: A powerful paradigm for scientific research","volume":"2","author":"Xu","year":"2021","journal-title":"Innovation"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.neunet.2014.09.003","article-title":"Deep learning in neural networks: An overview","volume":"61","author":"Schmidhuber","year":"2015","journal-title":"Neural Netw."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1186\/s40537-021-00444-8","article-title":"Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions","volume":"8","author":"Alzubaidi","year":"2021","journal-title":"J. Big Data"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Miao, J., Wang, Y., Ma, R., Liu, Y., Zhang, C., Zhang, J., Yu, Z., and Zheng, Y. (2023). Image recognition of traditional Chinese medicine based on deep learning. Front. Bioeng. Biotechnol., 11.","DOI":"10.3389\/fbioe.2023.1199803"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"304","DOI":"10.1080\/21642583.2021.1901159","article-title":"A novel Chinese herbal medicine classification approach based on EfficientNet","volume":"9","author":"Hao","year":"2021","journal-title":"Syst. Sci. Control Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"4604","DOI":"10.3390\/agriengineering6040263","article-title":"Deep learning-based model for effective classification of Ziziphus jujuba using RGB images","volume":"6","author":"Jeon","year":"2024","journal-title":"AgriEngineering"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Park, S.J., Lee, H., Jeon, Y.J., Woo, D.H., Kim, H.Y., Kim, J.O., Kim, H.-Y., and Jung, D.-H. (2025). Development of an RGB-GE data generation and XAI-based on-site classification system for differentiating Zizyphus jujuba and Zizyphus mauritiana in herbal medicine applications. Agriculture, 15.","DOI":"10.3390\/agriculture15101022"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Mulugeta, A.K., Sharma, D.P., and Mesfin, A.H. (2024). Deep learning for medicinal plant species classification and recognition: A systematic review. Front. Plant Sci., 14.","DOI":"10.3389\/fpls.2023.1286088"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Lin, M., Weng, N., Mikolaj, K., Bashir, Z., Svendsen, M.B.S., Tolsgaard, M.G., Feragen, A., and Christensen, A.N. (2024). Shortcut learning in medical image segmentation. Proceedings of the Medical Image Computing and Computer Assisted Intervention\u2014MICCAI 2024, Marrakesh, Morocco, 6\u201310 October 2024, Springer.","DOI":"10.1007\/978-3-031-72111-3_59"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1038\/s42256-020-00257-z","article-title":"Shortcut learning in deep neural networks","volume":"2","author":"Geirhos","year":"2020","journal-title":"Nat. Mach. Intell."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Kar, A., Tulsiani, S., Carreira, J., and Malik, J. (2015, January 7\u201313). Amodal completion and size constancy in natural scenes. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.","DOI":"10.1109\/ICCV.2015.23"},{"key":"ref_20","unstructured":"Hermann, K.L., Mobahi, H., Fel, T., and Mozer, M.C. (2024, January 7\u201311). On the foundations of shortcut learning. Proceedings of the 12th International Conference on Learning Representations (ICLR), Vienna, Austria."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Gilg, J., Teepe, T., Herzog, F., and Rigoll, G. (2023, January 3\u20137). The box size confidence bias harms your object detector. Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA.","DOI":"10.1109\/WACV56688.2023.00152"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"460","DOI":"10.1021\/acs.jnatprod.2c00929","article-title":"Botanical Ingredient Forensics: Detection of Attempts to Deceive Commonly Used Analytical Methods for Authenticating Herbal Dietary and Food Ingredients and Supplements","volume":"86","author":"Gafner","year":"2023","journal-title":"J. Nat. Prod."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"4314","DOI":"10.1038\/s41467-023-39902-7","article-title":"Detecting shortcut learning for fair medical AI using shortcut testing","volume":"14","author":"Brown","year":"2023","journal-title":"Nat. Commun."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"28689","DOI":"10.1007\/s11042-022-12935-8","article-title":"Deep learning based approach for digitized herbarium specimen segmentation","volume":"81","author":"Triki","year":"2022","journal-title":"Multimed. Tools Appl."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Chen, Y., Wu, Z., Zhao, B., Fan, C., and Shi, S. (2021). Weed and Corn Seedling Detection in Field Based on Multi Feature Fusion and Support Vector Machine. Sensors, 21.","DOI":"10.3390\/s21010212"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Doll\u00e1r, P. (2017, January 22\u201329). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref_27","unstructured":"Vandenhirtz, M., Manduchi, L., Marcinkevi\u010ds, R., and Vogt, J.E. (2023). Signal is harder to learn than bias: Debiasing with focal loss. arXiv."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Cui, Y., Jia, M., Lin, T.-Y., Song, Y., and Belongie, S. (2019, January 16\u201320). Class-balanced loss based on effective number of samples. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00949"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., and Wojna, Z. (2016, January 27\u201330). Rethinking the Inception architecture for computer vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Chattopadhyay, A., Sarkar, A., Howlader, P., and Balasubramanian, V.N. (2018, January 12\u201315). Grad-CAM++: Generalized gradient-based visual explanations for deep convolutional networks. Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, NV, USA.","DOI":"10.1109\/WACV.2018.00097"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1084","DOI":"10.1007\/s11263-017-1059-x","article-title":"Top-down neural attention by excitation backprop","volume":"126","author":"Zhang","year":"2018","journal-title":"Int. J. Comput. Vis."}],"container-title":["Biosensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-6374\/16\/6\/320\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T11:58:37Z","timestamp":1780401517000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-6374\/16\/6\/320"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,2]]},"references-count":32,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["bios16060320"],"URL":"https:\/\/doi.org\/10.3390\/bios16060320","relation":{},"ISSN":["2079-6374"],"issn-type":[{"value":"2079-6374","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,2]]}}}