ObjectiveTo investigate the screening value of cervical fluid-based cytology test (TCT), high-risk human papillomavirus (HR-HPV) test, and colposcopy for cervical intraepithelial neoplasia (CIN) and cervical cancer in high-risk populations.
MethodsA total of 466 patients between January 2013 and January 2015 with a history of intercourse bleeding were enrolled in this study, and the screening value of TCT, HR-HPV test and colposcopy for CIN and cervical cancer were retrospectively evaluated.
ResultsIn the 466 patients, 165 were diagnosed with cervical inflammation, 116 with CIN, 182 with grade 2-3 CIN, and 3 with cervical cancer. The colposcopy had the highest sensitivity (84.1%), the lowest specificity (59.4%), high false positive rate (40.6%), low false negative rate (15.9%), and the highest negative predictive value (67.1%). The TCT had the highest specificity (84.8%) and the lowest false positive rate (15.2%). The indicators of HR-HPV were between those of TCT and colposcopy. There were significant differences in terms of these indicators among the three methods (P < 0.05). And the positive prediction value of HR-HPV was the highest (84.5%), while the negative prediction value of colposcopy was the highest (67.1%). There was a significant difference in the predictive value among the three methods (P < 0.05). The consistency of either TCT or HR-HPV alone with pathological diagnosis was poor (K=0.213, 0.343), while that of colposcopy was moderate (K=0.446). Combination of TCT and HR-HPV could significantly improve the diagnosis sensitivity (93.0%) with a lower false negative rate (7.0%); Youden index was 0.736, and the consistency with pathological examination was high (K=0.748).
ConclusionsFor high-risk population with a history of intercourse bleeding or other abnormal cervical disorders, the screening sensitivity of TCT and HR-HPV alone for CIN and cervical cancer is low with a high false negative rate. Colposcopy has a high sensitivity and a low specificity. By combination of TCT and HR-HPV, the validity, reliability and predictive values can be improved significantly, and the sensitivity is high with a low false negative rate and a high consistency with pathological examination.
Cervical intraepithelial neoplasia is the primary type of cervical precancerous lesion; however, manual clinical diagnosis is prone to bias and has limited grading accuracy. To achieve precise automated grading of CIN, this paper proposes a multimodal fusion Swin Transformer model and develops a corresponding computer-aided diagnosis system. This method employs three-channel fusion of raw images, cervical mask images, and directional gradient histogram features to enhance lesion texture and location information. Within the Swin Transformer backbone, an atrous spatial pyramid pooling module channel attention module and a convolutional feature extraction module are embedded to balance global semantic and local detail features. A focal loss function is adopted to address class imbalance in the dataset and improve the model’s ability to identify difficult-to-classify samples. On a dataset of 3 915 clinical colposcopy images, the model achieved an overall accuracy of 90.01%, precision of 87.55%, recall of 86.17%, F1 score of 89.13%, outperforming baseline models such as VGG, ResNet, and Swin Transformer. The developed system integrates image quality screening, lesion identification, and three-level classification functions, providing an effective tool for the rapid and objective screening of clinical cervical precancerous lesions.
