Cervical cancer ranks as the fourth most prevalent cancer among women worldwide. Early diagnosis facilitates timely intervention and treatment. Traditional colposcopy is a widely employed technique for evaluating cervical lesions. Nevertheless, the growing number of patients with cervical cancer, and the rising workload of doctors may lead to misdiagnosis and underdiagnosis. Therefore, many researchers use deep learning to make paramedical diagnoses of cervical disease. However, the current neural network models for cervical image recognition lead to poor diagnostic results due to limitations in feature extraction. Consequently, this study proposes a novel model for classifying cervical images into normal, cervical intraepithelial neoplasia, and cancerous categories. The model proposes three new modules, namely a Feature Cognitive Screening Module, Multi-scale Feature Classification Module, and Overlapping Sampling Module (FSMO), which can realize the extraction of global features and local feature areas, multi-scale feature fusion classification and short-range interactions in the images, and enhance the edge capturing ability of the model and the proficiency of solving complex problems, thereby elevating prediction accuracy. The results showed that the recognition accuracy of FSMO in the self-constructed cervical image dataset was 91.88%, a precision of 92.91%, a recall of 91.92%, and an F1-Score of 91.99%; and the accuracy is 97.5% under the kaggle dataset. The performance surpasses other advanced models. Consequently, this model holds significant potential for rapid auxiliary diagnosis in cervical imaging, contributing to the early detection and treatment of cervical cancer.
BroutetNJeronimoJKumarS, et al. Implementation research to accelerate scale-up of national screen and treat strategies towards the elimination of cervical cancer. Prev Med2022; 155: 106906.
2.
GaoJAoYWangS, et al. WHO histological classification and tumor size are predictors of the locally aggressive behavior of thymic epithelial tumors. Lung Cancer2024; 187: 107446.
3.
BravoCAWalkerMJPapadopoulosA, et al. Social media use in HPV-, cervical cancer-, and cervical screening-related research: a scoping review. Prev Med2024; 179: 107798.
4.
PetryKUNieminenPJLeesonSC, et al. 2017 update of the European Federation for Colposcopy (EFC) performance standards for the practice of colposcopy. Eur J Obstet Gynecol Reprod Biol2018; 224: 137–141.
5.
KyrgiouMTsoumpouIVrekoussisT, et al. The up-to-date evidence on colposcopy practice and treatment of cervical intraepithelial neoplasia: the Cochrane colposcopy & cervical cytopathology collaborative group (C5 group) approach. Cancer Treat Rev2006; 32: 516–523.
6.
HosainMTJimJRMridhaMF, et al. Explainable AI approaches in deep learning: advancements, applications and challenges. Comput Electr Eng2024; 117: 109246.
7.
AbdullahMAbrhaFBKedirB, et al. A hybrid deep learning CNN model for COVID-19 detection from chest X-rays. Heliyon2024; 10: e26938.
8.
ZhangTLuoYMLiP, et al. Cervical precancerous lesions classification using pre-trained densely connected convolutional networks with colposcopy images. Biomed Signal Process Control2020; 55: 101566.
9.
KelenyiBDomsaVTamasL. SAM-Net: self-attention based feature matching with spatial transformers and knowledge distillation. Expert Syst Appl2024; 242: 122804.
10.
DosovitskiyABeyerLKolesnikovA, et al. An image is worth 16x16 words: transformers for image recognition at scale. In: International conference on learning representations (ICLR 2021). 2021.
11.
ŞahinEÖzdemirDTemurtaşH. Multi-objective optimization of ViT architecture for efficient brain tumor classification. Biomed Signal Process Control2024; 91: 105938.
12.
AlMohimeedAShehataMEl-RashidyN, et al. ViT-PSO-SVM: cervical cancer predication based on integrating vision transformer with particle swarm optimization and support vector machine. Bioengineering2024; 11: 729.
HongZXiongJYangH, et al. Lightweight low-rank adaptation vision transformer framework for cervical cancer detection and cervix type classification. Bioengineering2024; 11: 468.
15.
DongBXueHLiY, et al. Classification and diagnosis of cervical lesions based on colposcopy images using deep fully convolutional networks: a man-machine comparison cohort study. Fundam Res2025; 5: 419–428.
16.
KhareSKBoothBBBlanes-VidalV, et al. An explainable attention model for cervical precancer risk classification using colposcopic images. Comput Methods Programs Biomed2025; 271: 108976.
17.
KyrgiouMBowdenSJAthanasiouA, et al. Morbidity after local excision of the transformation zone for cervical intra-epithelial neoplasia and early cervical cancer. Best Pract Res Clin Obstet Gynaecol2021; 75: 10–22.
18.
LitjensGKooiTBejnordiBE, et al. A survey on deep learning in medical image analysis. Med Image Anal2017; 42: 60–88.
19.
KurupARWigdahlJBensonJ, et al. Automated malarial retinopathy detection using transfer learning and multi-camera retinal images. Biocybern Biomed Eng2023; 43: 109–123.
20.
LopesUKValiatiJF. Pre-trained convolutional neural networks as feature extractors for tuberculosis detection. Comput Biol Med2017; 89: 135–143.
21.
LuXYeXChengY. An overlapping minimization-based over-sampling algorithm for binary imbalanced classification. Eng Appl Artif Intell2024; 133: 108107.
HouQXiaRZhangJ, et al. Learning visual overlapping image pairs for SfM via CNN fine-tuning with photogrammetric geometry information. Int J Appl Earth Obs Geoinf2023; 116: 103162.
24.
SunJWuBZhaoT, et al. Classification for thyroid nodule using ViT with contrastive learning in ultrasound images. Comput Biol Med2023; 152: 106444.
25.
ZhuDWangD. Transformers and their application to medical image processing: a review. J Radiat Res Appl Sci2023; 16: 100680.
26.
WangYDuBWangW, et al. Multi-tailed vision transformer for efficient inference. Neural Netw2024; 174: 106235.
27.
DarMFGanivadaA. Deep learning and genetic algorithm-based ensemble model for feature selection and classification of breast ultrasound images. Image Vis Comput2024; 146: 105018.
28.
HeoJSeoSKangP. Exploring the differences in adversarial robustness between ViT- and CNN-based models using novel metrics. Comput Vis Image Underst2023; 235: 103800.
29.
WuX-JXuS-KGaoM-Y, et al. A new ECT image reconstruction algorithm based on Vision transformer (ViT). Flow Meas Instrum2024; 97: 102611.
30.
ParvaizAKhalidMAZafarR, et al. Vision Transformers in medical computer vision—a contemplative retrospection. Eng Appl Artif Intell2023; 122: 106126.
31.
LiuCMengZ. TBFF-DAC: two-branch feature fusion based on deformable attention and convolution for object detection. Comput Electr Eng2024; 116: 109132.
32.
LiuSWangLYueW. An efficient medical image classification network based on multi-branch CNN, token grouping transformer and mixer MLP. Appl Soft Comput2024; 153: 111323.
33.
TouvronHCordMJégouH, et al. DeiT III: revenge of the ViT. In: AvidanSBrostowGCisséM (eds) Computer Vision – ECCV 2022. Springer Nature, 2022, pp. 516–533.
34.
SrinivasALinT-YParmarN, et al. Bottleneck transformers for visual recognition. In: 2021 IEEE/CVF conference on computer vision and pattern recognition (CVPR), Nashville, TN, 2021, pp. 16519–16529. IEEE.
35.
YuanLChenYWangT, et al. Tokens-to-token ViT: training vision transformers from scratch on ImageNet. In: 2021 IEEE/CVF international conference on computer vision (ICCV), Montreal, QC, Canada, 2021, pp. 558–567. IEEE.
36.
YangTHuHLiX, et al. An efficient fusion-purification network for cervical pap-smear image classification. Comput Methods Programs Biomed2024; 251: 108199.
37.
ZhangXZhaoS. Cervical image classification based on image segmentation preprocessing and a CapsNet network model. Int J Imaging Syst Technol2019; 29: 19–28.
38.
QinJHeYLiangY, et al. Cell comparative learning: A cervical cytopathology whole slide image classification method using normal and abnormal cells. Computerized Medical Imaging and Graphics2024; 117: 102427.
39.
WangQXiongYZhuH, et al. Cervical OCT image classification using contrastive masked autoencoders with Swin Transformer. Computerized Medical Imaging and Graphics2024; 118: 102469.
40.
YangTHuHLiX, et al. A pyramid convolutional mixer for cervical pap-smear image classification tasks. Biomed Signal Process Control2025; 99: 106789.
41.
LiuWLiCXuN, et al. CVM-Cervix: a hybrid cervical pap-smear image classification framework using CNN, visual transformer and multilayer perceptron. Pattern Recognit2022; 130: 108829.
