{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,15]],"date-time":"2025-08-15T00:20:48Z","timestamp":1755217248303,"version":"3.43.0"},"reference-count":60,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2025,8,5]],"date-time":"2025-08-05T00:00:00Z","timestamp":1754352000000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program","doi-asserted-by":"crossref","award":["2019YFC0117800"],"award-info":[{"award-number":["2019YFC0117800"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>Cervical cancer ranks first in incidence among malignant tumors of the female reproductive system, and 80% of women who die from cervical cancer worldwide are from developing countries. Visual inspection with acetic acid (VIA) screening based on artificial intelligence-assisted diagnosis can provide a cheap and rapid screening method. This will attract more low-income women to volunteer for regular cervical cancer screening. However, current AI-based VIA screening studies either have low accuracy or require expensive equipment assistance. In this article, we propose the Hierarchical Multi-Scale Convolutional Transformer network, which combines the hierarchical feature extraction capability of Convolutional Neural Network (CNNs) and the global dependency modeling capability of Transformers to address the challenges of realizing intelligent VIA screening. Hierarchical multi-scale convolutional transformer (HMCFormer) can be divided into a Transformer branch and a CNN branch. The Transformer branch receives unenhanced lesion sample images, and the CNN branch receives lesion sample images enhanced by the proposed dual-color space-based image enhancement algorithm. The authors design a hierarchical multi-scale pixel excitation module for adaptive multi-scale and multi-level local feature extraction. The authors apply the structure of the Swin Transformer network with minor modifications in the global perception modeling process. In addition, the authors propose two feature fusion concepts: adaptive preprocessing and superiority-inferiority fusion, and design a feature fusion module based on these concepts, which significantly improves the collaborative ability of the Transformer branch and the CNN branch. The authors collected and summarized 5,000 samples suitable for VIA screening methods from public datasets provided by companies such as Intel and Google, forming the PCC5000 dataset. On this dataset, the proposed algorithm achieves a screening accuracy of 97.4% and a grading accuracy of 94.8%.<\/jats:p>","DOI":"10.7717\/peerj-cs.3088","type":"journal-article","created":{"date-parts":[[2025,8,5]],"date-time":"2025-08-05T08:46:26Z","timestamp":1754383586000},"page":"e3088","source":"Crossref","is-referenced-by-count":0,"title":["HMCFormer (hierarchical multi-scale convolutional transformer): a hybrid CNN+Transformer network for intelligent VIA screening"],"prefix":"10.7717","volume":"11","author":[{"given":"Bo","family":"Feng","sequence":"first","affiliation":[{"name":"School of Integrated Ciruits, Anhui University, HeFei, Anhui, China"},{"name":"Anhui Engineering Laboratory of Agro-Ecological Big Data, Anhui University, HeFei, Anhui, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chao","family":"Xu","sequence":"additional","affiliation":[{"name":"School of Integrated Ciruits, Anhui University, HeFei, Anhui, China"},{"name":"Anhui Engineering Laboratory of Agro-Ecological Big Data, Anhui University, HeFei, Anhui, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhengping","family":"Li","sequence":"additional","affiliation":[{"name":"School of Integrated Ciruits, Anhui University, HeFei, Anhui, China"},{"name":"Anhui Engineering Laboratory of Agro-Ecological Big Data, Anhui University, HeFei, Anhui, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chuanyi","family":"Zhang","sequence":"additional","affiliation":[{"name":"Obstetrics and Gynecology, People\u2019s Hospital of Fanchang District, Wuhu, Anhui, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2025,8,5]]},"reference":[{"issue":"7","key":"10.7717\/peerj-cs.3088\/ref-1","doi-asserted-by":"publisher","first-page":"1620","DOI":"10.3390\/agronomy12071620","article-title":"Precision chemical weed management strategies: a review and a design of a new CNN-based modular spot sprayer","volume":"12","author":"Allmendinger","year":"2022","journal-title":"Agronomy"},{"key":"10.7717\/peerj-cs.3088\/ref-2","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2308.13917","article-title":"Transfer learning for microstructure segmentation with CS-UNet: a hybrid algorithm with transformer and CNN encoders","author":"Alrfou","year":"2023"},{"issue":"14","key":"10.7717\/peerj-cs.3088\/ref-3","doi-asserted-by":"publisher","first-page":"5464","DOI":"10.3390\/s22145464","article-title":"EIEN: endoscopic image enhancement network based on retinex theory","volume":"22","author":"An","year":"2022","journal-title":"Sensors"},{"issue":"8","key":"10.7717\/peerj-cs.3088\/ref-4","doi-asserted-by":"publisher","first-page":"2306","DOI":"10.1101\/324541","article-title":"Development of algorithms for automated detection of cervical pre-cancers with a low-cost, point-of-care, pocket colposcope","volume":"66","author":"Asiedu","year":"2018","journal-title":"IEEE Transactions on Biomedical Engineering"},{"key":"10.7717\/peerj-cs.3088\/ref-5","first-page":"11142","article-title":"Conformer: local features coupling global representations for visual recognition","author":"Chen","year":"2021"},{"key":"10.7717\/peerj-cs.3088\/ref-6","first-page":"5710","article-title":"LEFormer: a hybrid CNN-transformer architecture for accurate lake extraction from remote sensing imagery","author":"Chen","year":"2024"},{"key":"10.7717\/peerj-cs.3088\/ref-7","first-page":"2286","article-title":"Improving vision transformers with soft convolutional inductive biases","author":"d\u2019Ascoli","year":"2021"},{"key":"10.7717\/peerj-cs.3088\/ref-8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2402.11301","article-title":"ReViT: enhancing vision transformers with attention residual connections for visual recognition","author":"Diko","year":"2024"},{"issue":"10","key":"10.7717\/peerj-cs.3088\/ref-9","doi-asserted-by":"publisher","first-page":"9675628","DOI":"10.1155\/2022\/9675628","article-title":"An improved image classification method for cervical precancerous lesions based on ShuffleNet","volume":"2022","author":"Fang","year":"2022","journal-title":"Computational Intelligence and Neuroscience"},{"issue":"7","key":"10.7717\/peerj-cs.3088\/ref-10","doi-asserted-by":"publisher","first-page":"1246","DOI":"10.3390\/electronics14071246","article-title":"WLAM attention: plug-and-play wavelet transform linear attention","volume":"14","author":"Feng","year":"2025","journal-title":"Electronics (2079\u20139292)"},{"issue":"9","key":"10.7717\/peerj-cs.3088\/ref-11","first-page":"815","article-title":"A comparison of single and combined visual cytologic and virologic tests as screening strategies in a region at high risk of cervical cancer","volume":"12","author":"Ferreccio","year":"2003","journal-title":"Cancer Epidemiology"},{"issue":"2","key":"10.7717\/peerj-cs.3088\/ref-12","doi-asserted-by":"publisher","first-page":"652","DOI":"10.1109\/TPAMI.2019.2938758","article-title":"Res2Net: a new multi-scale backbone architecture","volume":"43","author":"Gao","year":"2019","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.7717\/peerj-cs.3088\/ref-13","doi-asserted-by":"publisher","first-page":"1978","DOI":"10.1109\/TIP.2023.3261747","article-title":"CTCNet: a CNN-transformer cooperation network for face image super-resolution","volume":"32","author":"Gao","year":"2023","journal-title":"IEEE Transactions on Image Processing"},{"key":"10.7717\/peerj-cs.3088\/ref-14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2405.16605","article-title":"Demystify Mamba in vision: a linear attention perspective","author":"Han","year":"2024"},{"key":"10.7717\/peerj-cs.3088\/ref-15","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.08874","article-title":"Agent attention: on the integration of softmax and linear attention","author":"Han","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-16","doi-asserted-by":"publisher","DOI":"10.1101\/2020.08.12.20172346","article-title":"Point-of-care digital cytology with artificial intelligence for cervical cancer screening at a peripheral clinic in Kenya","author":"Holmstr\u00f6m","year":"2020"},{"issue":"9","key":"10.7717\/peerj-cs.3088\/ref-17","doi-asserted-by":"publisher","first-page":"923","DOI":"10.1093\/jnci\/djy225","article-title":"An observational study of deep learning and automated evaluation of cervical images for cancer screening","volume":"111","author":"Hu","year":"2019","journal-title":"JNCI: Journal of the National Cancer Institute"},{"issue":"4","key":"10.7717\/peerj-cs.3088\/ref-18","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1097\/lgt.0000000000000338","article-title":"ASCCP colposcopy standards: role of colposcopy, benefits, potential harms, and terminology for colposcopic practice","volume":"21","author":"Khan","year":"2017","journal-title":"Journal of Lower Genital Tract Disease"},{"key":"10.7717\/peerj-cs.3088\/ref-19","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1007\/978-1-4842-6168-2_11","volume-title":"Convolutional neural networks with swift for Tensorflow: image recognition and dataset categorization","author":"Koonce","year":"2021"},{"issue":"11","key":"10.7717\/peerj-cs.3088\/ref-20","doi-asserted-by":"publisher","first-page":"3403","DOI":"10.1109\/tmi.2020.2994778","article-title":"Computer-aided cervical cancer diagnosis using time-lapsed colposcopic images","volume":"39","author":"Li","year":"2020","journal-title":"IEEE Transactions on Medical Imaging"},{"key":"10.7717\/peerj-cs.3088\/ref-21","doi-asserted-by":"publisher","first-page":"12934","DOI":"10.48550\/arXiv.2206.01191","article-title":"EfficientFormer: vision transformers at MobileNet speed","volume":"35","author":"Li","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"6","key":"10.7717\/peerj-cs.3088\/ref-22","doi-asserted-by":"publisher","first-page":"2254","DOI":"10.1109\/TMI.2024.3363190","article-title":"Transformer makes CNN work better for scribble-based medical image segmentation","volume":"43","author":"Li","year":"2024","journal-title":"IEEE Transactions on Medical Imaging"},{"issue":"13","key":"10.7717\/peerj-cs.3088\/ref-23","first-page":"1","article-title":"Application value of colposcopy combined with white acetate test in the cervical cancer screening","volume":"12","author":"Liu","year":"2014","journal-title":"Guide of China Medicine"},{"key":"10.7717\/peerj-cs.3088\/ref-24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2104.13497","article-title":"ConTNet: why not use convolution and transformer at the same time","author":"Liu","year":"2023a"},{"key":"10.7717\/peerj-cs.3088\/ref-25","first-page":"12009","article-title":"Swin Transformer V2: scaling up capacity and resolution","author":"Liu","year":"2022"},{"key":"10.7717\/peerj-cs.3088\/ref-26","first-page":"10012","article-title":"Swin Transformer: hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"key":"10.7717\/peerj-cs.3088\/ref-27","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2305.07027","article-title":"EfficientViT: memory efficient vision transformer with cascaded group attention (Version 1)","author":"Liu","year":"2023b"},{"key":"10.7717\/peerj-cs.3088\/ref-28","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.10166","article-title":"VMamba: visual state space model","author":"Liu","year":"2024"},{"key":"10.7717\/peerj-cs.3088\/ref-29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2310.19380","article-title":"TransXNet: learning both global and local dynamics with a dual dynamic token mixer for visual recognition","author":"Lou","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-30","doi-asserted-by":"publisher","first-page":"29616","DOI":"10.1109\/access.2020.2972610","article-title":"MDFI: multi-CNN decision feature integration for diagnosis of cervical precancerous lesions","volume":"8","author":"Luo","year":"2020","journal-title":"IEEE Access"},{"issue":"6","key":"10.7717\/peerj-cs.3088\/ref-31","doi-asserted-by":"publisher","first-page":"583","DOI":"10.3892\/mco.2019.1932","article-title":"Application of deep learning to the classification of uterine cervical squamous epithelial lesion from colposcopy images","volume":"11","author":"Miyagi","year":"2019","journal-title":"Molecular and Clinical Oncology"},{"issue":"1","key":"10.7717\/peerj-cs.3088\/ref-32","first-page":"85","article-title":"The white paper on the current status of cervical cancer screening and early precision diagnosis in 2024","volume":"40","author":"Qiu","year":"2024","journal-title":"Chinese Journal of Practical Gynecology and Obstetrics"},{"issue":"1","key":"10.7717\/peerj-cs.3088\/ref-33","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1258\/td.2009.090085","article-title":"Evaluation of visual inspection with acetic acid and Lugol\u2019s iodine as cervical cancer screening tools in a low-resource setting","volume":"40","author":"Qureshi","year":"2010","journal-title":"Tropical Doctor"},{"issue":"1","key":"10.7717\/peerj-cs.3088\/ref-34","doi-asserted-by":"publisher","first-page":"104649","DOI":"10.1016\/j.compbiomed.2021.104649","article-title":"DeepCervix: a deep learning-based framework for the classification of cervical cells using hybrid deep feature fusion techniques","volume":"136","author":"Rahaman","year":"2021","journal-title":"Computers in Biology and Medicine"},{"issue":"17","key":"10.7717\/peerj-cs.3088\/ref-35","doi-asserted-by":"publisher","first-page":"3689","DOI":"10.3390\/math11173689","article-title":"Cervical precancerous lesion image enhancement based on retinex and histogram equalization","volume":"11","author":"Ren","year":"2023","journal-title":"Mathematics"},{"issue":"31","key":"10.7717\/peerj-cs.3088\/ref-36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00138-020-01063-8","article-title":"ColpoNet for automated cervical cancer screening using colposcopy images","volume":"2020","author":"Saini","year":"2020","journal-title":"Machine Vision and Applications"},{"key":"10.7717\/peerj-cs.3088\/ref-37","first-page":"4510","article-title":"MobileNetV2: inverted residuals and linear bottlenecks","author":"Sandler","year":"2018"},{"issue":"6","key":"10.7717\/peerj-cs.3088\/ref-38","doi-asserted-by":"publisher","first-page":"907","DOI":"10.1002\/ijc.20190","article-title":"Accuracy of visual screening for cervical neoplasia: results from an IARC multicentre study in India and Africa","volume":"110","author":"Sankaranarayanan","year":"2004","journal-title":"International Journal of Cancer"},{"key":"10.7717\/peerj-cs.3088\/ref-39","first-page":"17425","article-title":"SwiftFormer: efficient additive attention for transformer-based real-time mobile vision applications","author":"Shaker","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-40","first-page":"2136","article-title":"TransMIL: transformer based correlated multiple instance learning for whole slide image classification","volume":"34","author":"Shao","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"1","key":"10.7717\/peerj-cs.3088\/ref-41","doi-asserted-by":"publisher","first-page":"62","DOI":"10.2174\/1389202920666190313163414","article-title":"A stack autoencoders based deep neural network approach for cervical cell classification in pap-smear images","volume":"14","author":"Singh","year":"2021","journal-title":"Recent Advances in Computer Science and Communications (Formerly: Recent Patents on Computer Science)"},{"issue":"1","key":"10.7717\/peerj-cs.3088\/ref-42","doi-asserted-by":"publisher","first-page":"229","DOI":"10.1109\/tmi.2014.2352311","article-title":"Multimodal entity coreference for cervical dysplasia diagnosis","volume":"34","author":"Song","year":"2014","journal-title":"IEEE Transactions on Medical Imaging"},{"article-title":"Rethinking the inception architecture for computer vision","year":"2024","author":"Szegedy","key":"10.7717\/peerj-cs.3088\/ref-43"},{"key":"10.7717\/peerj-cs.3088\/ref-44","first-page":"10096","article-title":"EfficientNetV2: smaller models and faster training","author":"Tan","year":"2021"},{"issue":"12","key":"10.7717\/peerj-cs.3088\/ref-45","doi-asserted-by":"publisher","first-page":"1909","DOI":"10.3390\/electronics11121909","article-title":"An endoscope image enhancement algorithm based on image decomposition","volume":"11","author":"Tan","year":"2022","journal-title":"Electronics"},{"issue":"9","key":"10.7717\/peerj-cs.3088\/ref-46","doi-asserted-by":"publisher","first-page":"693","DOI":"10.1002\/cncy.22425","article-title":"Cervical cytology screening facilitated by an artificial intelligence microscope: a preliminary study","volume":"129","author":"Tang","year":"2021","journal-title":"Cancer Cytopathology"},{"issue":"11","key":"10.7717\/peerj-cs.3088\/ref-47","doi-asserted-by":"publisher","first-page":"1293","DOI":"10.1111\/j.1471-0528.2012.03444.x","article-title":"Accuracy of colposcopy-directed punch biopsies: a systematic review and meta-analysis: systematic review of the accuracy of punch biopsies","volume":"119","author":"Underwood","year":"2012","journal-title":"BJOG: An International Journal of Obstetrics & Gynaecology"},{"key":"10.7717\/peerj-cs.3088\/ref-48","first-page":"5785","article-title":"FastViT: a fast hybrid vision transformer using structural reparameterization","author":"Vasu","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-49","first-page":"390","article-title":"CSPNet: a new backbone that can enhance learning capability of CNN","author":"Wang","year":"2020"},{"key":"10.7717\/peerj-cs.3088\/ref-50","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tim.2024.3379090","article-title":"YOLOv8-QSD: an improved small object detection algorithm for autonomous vehicles based on YOLOv8","volume":"73","author":"Wang","year":"2024","journal-title":"IEEE Transactions on Instrumentation and Measurement"},{"issue":"3","key":"10.7717\/peerj-cs.3088\/ref-51","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1007\/s41095-022-0274-8","article-title":"PVT v2: improved baselines with pyramid vision transformer","volume":"8","author":"Wang","year":"2022","journal-title":"Computational Visual Media"},{"key":"10.7717\/peerj-cs.3088\/ref-52","first-page":"16133","article-title":"ConvNeXt V2: co-designing and scaling ConvNets with masked autoencoders","author":"Woo","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-53","first-page":"171","article-title":"CoTr: efficiently bridging CNN and transformer for 3D medical image segmentation","author":"Xie","year":"2021"},{"key":"10.7717\/peerj-cs.3088\/ref-54","doi-asserted-by":"publisher","first-page":"4202","DOI":"10.1109\/TIP.2024.3425048","article-title":"HAFormer: unleashing the power of hierarchy-aware features for lightweight semantic segmentation","volume":"33","author":"Xu","year":"2024","journal-title":"IEEE Transactions on Image Processing"},{"key":"10.7717\/peerj-cs.3088\/ref-55","first-page":"4956","article-title":"Enriched CNN-transformer feature aggregation networks for super-resolution","author":"Yoo","year":"2023"},{"key":"10.7717\/peerj-cs.3088\/ref-56","first-page":"10819","article-title":"MetaFormer is actually what you need for vision","author":"Yu","year":"2022"},{"key":"10.7717\/peerj-cs.3088\/ref-57","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2408.03703","article-title":"CAS-ViT: convolutional additive self-attention vision transformers for efficient mobile applications","author":"Zhang","year":"2024"},{"issue":"6","key":"10.7717\/peerj-cs.3088\/ref-58","doi-asserted-by":"publisher","first-page":"1633","DOI":"10.1109\/jbhi.2017.2705583","article-title":"DeepPap: deep convolutional networks for cervical cell classification","volume":"21","author":"Zhang","year":"2017","journal-title":"IEEE Journal of Biomedical and Health Informatics"},{"key":"10.7717\/peerj-cs.3088\/ref-59","doi-asserted-by":"publisher","first-page":"101566","DOI":"10.1016\/j.bspc.2019.101566","article-title":"Cervical precancerous lesions classification using pre-trained densely connected convolutional networks with colposcopy images","volume":"55","author":"Zhang","year":"2020","journal-title":"Biomedical Signal Processing and Control"},{"key":"10.7717\/peerj-cs.3088\/ref-60","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2204.03055","article-title":"Mobile-Former: bridging MobileNet and transformer for efficient image recognition","author":"Zhu","year":"2022"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3088.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3088.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3088.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3088.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,5]],"date-time":"2025-08-05T08:46:31Z","timestamp":1754383591000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3088"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,5]]},"references-count":60,"alternative-id":["10.7717\/peerj-cs.3088"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3088","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"type":"electronic","value":"2376-5992"}],"subject":[],"published":{"date-parts":[[2025,8,5]]},"article-number":"e3088"}}