{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,20]],"date-time":"2026-10-20T05:55:32Z","timestamp":1792475732463,"version":"4.5.0"},"reference-count":27,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T00:00:00Z","timestamp":1773878400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100003052","name":"Ministry of Trade, Industry and Energy","doi-asserted-by":"publisher","award":["20224000000090"],"award-info":[{"award-number":["20224000000090"]}],"id":[{"id":"10.13039\/501100003052","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003565","name":"Ministry of Land, Infrastructure and Transport","doi-asserted-by":"publisher","award":["Innovative Talent Education Program for Smart City"],"award-info":[{"award-number":["Innovative Talent Education Program for Smart City"]}],"id":[{"id":"10.13039\/501100003565","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    Accurate identification of toxic Chinese herbal medicines is critical for public health and clinical safety. However, real-world herbal images often exhibit complex backgrounds and small, indistinct target regions, posing substantial challenges to automated classification systems. In this study, we present a novel image dataset comprising over 4,000 samples from 47 toxic herb categories, captured under diverse environmental conditions to reflect real-world variability. We benchmark several state-of-the-art convolutional neural networks, including ResNet, ResNeXt, and EfficientNet variants, and identify EfficientNetV2 as the most effective baseline. To further enhance model robustness and discriminative capability, we propose an improved EfficientNetV2 architecture incorporating two lightweight yet effective modules: a Multi-Scale Feature Fusion (MSFF) module to integrate hierarchical features, and a Convolutional Block Attention Module (CBAM) to refine both spatial and channel-wise representations. Experimental results demonstrate that our enhanced model achieves 91.28% Top-1 accuracy, 97.52% Top-5 accuracy, and a 90.27% macro F1-score, significantly outperforming baseline methods. Ablation studies confirm the complementary benefits of MSFF and CBAM, and targeted evaluations on challenging image subsets reveal improved resilience to background clutter and small object localization. The proposed architecture offers a high-accuracy, generalizable, and computationally efficient solution for toxic herbal medicine classification and provides a valuable reference for intelligent traditional medicine recognition applications. The GitHub repository for this project is available at:\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/zhuguohua1992\/Toxic-chinese-herbal-medicine-recognition-via-enhanced-EfficientNetV2\" xlink:type=\"simple\">https:\/\/github.com\/zhuguohua1992\/Toxic-chinese-herbal-medicine-recognition-via-enhanced-EfficientNetV2<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0344262","type":"journal-article","created":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T20:02:17Z","timestamp":1773950537000},"page":"e0344262","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":2,"title":["Toxic chinese herbal medicine recognition in real-world images via multi-scale and attention-enhanced EfficientNetV2"],"prefix":"10.1371","volume":"21","author":[{"given":"Guohua","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaehan","family":"Joo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Seonghyeon","family":"Park","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5699-8565","authenticated-orcid":true,"given":"Suk Chan","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,3,19]]},"reference":[{"key":"pone.0344262.ref001","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1016\/bs.irn.2017.02.003","article-title":"Effect of Chinese Herbal Medicine on Alzheimer\u2019s Disease","volume":"135","author":"M-JR Howes","year":"2017","journal-title":"Int Rev Neurobiol"},{"issue":"3","key":"pone.0344262.ref002","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1016\/j.jep.2006.01.010","article-title":"Cosmetic applications of selected traditional Chinese herbal medicines","volume":"106","author":"K-H Wang","year":"2006","journal-title":"J Ethnopharmacol"},{"issue":"9","key":"pone.0344262.ref003","doi-asserted-by":"crossref","first-page":"1223","DOI":"10.3109\/13880209.2014.884606","article-title":"Chinese herbal medicine for bone health","volume":"52","author":"E Mukwaya","year":"2014","journal-title":"Pharm Biol"},{"key":"pone.0344262.ref004","doi-asserted-by":"crossref","first-page":"301","DOI":"10.1016\/bs.apha.2019.08.001","article-title":"Toxicity of traditional Chinese medicine herbal and mineral products","volume":"87","author":"R Liu","year":"2020","journal-title":"Adv Pharmacol"},{"issue":"4","key":"pone.0344262.ref005","doi-asserted-by":"crossref","first-page":"215","DOI":"10.1016\/j.toxac.2022.04.003","article-title":"A review of botanical, biogeographical phytochemical and toxicological aspects of the toxic plants in Morocco","volume":"34","author":"M Elouardi","year":"2022","journal-title":"Toxicologie Analytique et Clinique"},{"key":"pone.0344262.ref006","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1016\/j.neucom.2013.02.014","article-title":"New image descriptors based on color, texture, shape, and wavelets for object and scene image classification","volume":"117","author":"S Banerji","year":"2013","journal-title":"Neurocomputing"},{"issue":"1","key":"pone.0344262.ref007","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1016\/j.aci.2016.04.001","article-title":"Ensemble of texture descriptors and classifiers for face recognition","volume":"13","author":"A Lumini","year":"2017","journal-title":"Applied Computing and Informatics"},{"issue":"2","key":"pone.0344262.ref008","doi-asserted-by":"crossref","first-page":"160","DOI":"10.1504\/IJAPR.2015.069538","article-title":"Fruit and vegetable recognition by fusing colour and texture features of the image using machine learning","volume":"2","author":"SR Dubey","year":"2015","journal-title":"IJAPR"},{"issue":"3","key":"pone.0344262.ref009","doi-asserted-by":"crossref","first-page":"1997","DOI":"10.3233\/JIFS-169911","article-title":"A study on plant recognition using conventional image processing and deep learning approaches","volume":"36","author":"S Anubha Pearline","year":"2019","journal-title":"IFS"},{"key":"pone.0344262.ref010","doi-asserted-by":"crossref","unstructured":"Yalcin H, Razavi S. Plant classification using convolutional neural networks. In: 2016 Fifth International Conference on Agro-Geoinformatics (Agro-Geoinformatics), 2016. 1\u20135. https:\/\/doi.org\/10.1109\/agro-geoinformatics.2016.7577698","DOI":"10.1109\/Agro-Geoinformatics.2016.7577698"},{"key":"pone.0344262.ref011","doi-asserted-by":"crossref","unstructured":"Wu N, Lou J, Lv J, Liu F, Wu X. Chinese herbal recognition by Spatial-\/Channel-wise attention. In: 2022 IEEE Conference on Telecommunications, Optics and Computer Science (TOCS), 2022. 1145\u201350. https:\/\/doi.org\/10.1109\/tocs56154.2022.10016159","DOI":"10.1109\/TOCS56154.2022.10016159"},{"issue":"6","key":"pone.0344262.ref012","article-title":"Chinese Herbal Medicine Image Recognition and Retrieval by Convolutional Neural Network","volume":"11","author":"X Sun","year":"2016","journal-title":"PLoS One"},{"key":"pone.0344262.ref013","first-page":"7759","article-title":"Research on micro-morphological feature recognition method of traditional Chinese medicinal herbs based on deep convolutional neural network","volume":"127","author":"H Wang","year":"2025","journal-title":"Journal of Combinatorial Mathematics and Combinatorial Computing"},{"key":"pone.0344262.ref014","doi-asserted-by":"crossref","first-page":"549","DOI":"10.1201\/9781003357346-61","article-title":"Herbal Plants Leaf Image Classification Using Machine Learning Approach","volume-title":"Intelligent Systems and Smart Infrastructure","author":"G Kumar","year":"2023"},{"issue":"15","key":"pone.0344262.ref015","first-page":"1952","article-title":"Automated Real-Time Identification of Medicinal Plants Species in Natural Environment Using Deep Learning Models-A Case Study from Borneo Region","volume":"11","author":"OA Malik","year":"2022","journal-title":"Plants (Basel)"},{"key":"pone.0344262.ref016","doi-asserted-by":"crossref","first-page":"100607","DOI":"10.1016\/j.imu.2021.100607","article-title":"An easy method for identifying 315 categories of commonly-used Chinese herbal medicines based on automated image recognition using AutoML platforms","volume":"25","author":"W Chen","year":"2021","journal-title":"Informatics in Medicine Unlocked"},{"key":"pone.0344262.ref017","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. 770\u20138. https:\/\/doi.org\/10.1109\/cvpr.2016.90","DOI":"10.1109\/CVPR.2016.90"},{"key":"pone.0344262.ref018","doi-asserted-by":"crossref","unstructured":"Xie S, Girshick R, Dollar P, Tu Z, He K. Aggregated Residual Transformations for Deep Neural Networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. 5987\u201395. https:\/\/doi.org\/10.1109\/cvpr.2017.634","DOI":"10.1109\/CVPR.2017.634"},{"key":"pone.0344262.ref019","first-page":"109","volume-title":"EfficientNet. In: Convolutional Neural Networks with Swift for TensorFlow: Image Recognition and Dataset Categorization","author":"B Koonce","year":"2021"},{"key":"pone.0344262.ref020","unstructured":"Tan M, Le Q. EfficientNetV2: Smaller models and faster training. In: Proceedings of the International Conference on Machine Learning, 2021. 10096\u2013106."},{"key":"pone.0344262.ref021","doi-asserted-by":"crossref","first-page":"881021","DOI":"10.3389\/fnbot.2022.881021","article-title":"Multi-Scale Feature Fusion Convolutional Neural Network for Indoor Small Target Detection","volume":"16","author":"L Huang","year":"2022","journal-title":"Front Neurorobot"},{"key":"pone.0344262.ref022","first-page":"3","article-title":"CBAM: Convolutional Block Attention Module","volume-title":"Lecture Notes in Computer Science","author":"S Woo","year":"2018"},{"key":"pone.0344262.ref023","doi-asserted-by":"crossref","first-page":"123893","DOI":"10.1016\/j.eswa.2024.123893","article-title":"Exploring Multiple Instance Learning (MIL): A brief survey","volume":"250","author":"M Waqas","year":"2024","journal-title":"Expert Systems with Applications"},{"key":"pone.0344262.ref024","doi-asserted-by":"crossref","first-page":"119113","DOI":"10.1016\/j.eswa.2022.119113","article-title":"Robust bag classification approach for multi-instance learning via subspace fuzzy clustering","volume":"214","author":"M Waqas","year":"2023","journal-title":"Expert Systems with Applications"},{"key":"pone.0344262.ref025","article-title":"From classical machine learning to emerging foundation models: Review on multimodal data integration for cancer research","author":"A Muneer","year":"2025","journal-title":"arXiv preprint"},{"key":"pone.0344262.ref026","unstructured":"Dosovitskiy A. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint. 2020. arXiv:2010.11929."},{"key":"pone.0344262.ref027","doi-asserted-by":"crossref","unstructured":"Liu Z, Lin Y, Cao Y. Swin Transformer: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE\/CVF International Conference on Computer Vision, 2021. 10012\u201322.","DOI":"10.1109\/ICCV48922.2021.00986"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344262","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T20:02:23Z","timestamp":1773950543000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344262"}},"subtitle":[],"editor":[{"given":"Amgad","family":"Muneer","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,19]]},"references-count":27,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,19]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0344262","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,19]]}}}