AI in Clinical Medicine, ISSN 2819-7437 online, Open Access
Article copyright, the authors; Journal compilation copyright, AI Clin Med and Elmer Press Inc
Journal website https://aicm.elmerpub.com

Review

Volume 2, March 2026, e16

Artificial Intelligence in Colposcopy: Diagnostic Performance, Clinical Applications, and Future Directions

Figures

Figure 1.
Figure 1. PRISMA flow diagram.
Figure 2.
Figure 2. Artificial intelligence (AI)-assisted colposcopy workflow.
Figure 3.
Figure 3. Multi-panel synthesis of diagnostic performance and functional roles of artificial intelligence (AI) in colposcopy. (a) Summary receiver operating characteristic (ROC) plot illustrating sensitivity and specificity trade-offs across representative studies evaluating AI-assisted detection of high-grade cervical lesions. Each point represents a study. (b) Flowchart illustrating major clinical task categories of AI in colposcopy, including detection, classification, triage, and clinical decision support.

Table

Table 1. Summary of Key Studies Evaluating Artificial Intelligence–Assisted Colposcopy
 
StudyYearStudy designDatasetAI methodTarget outcomePerformance
AUC: area under the receiver operating characteristic curve; CNN: convolutional neural network.
Hu et al [5]2019Retrospective9,406 imagesCNNCIN2+ vs. ≤ CIN1Sensitivity 94.0%, specificity 88.0%, AUC 0.96
Song et al [6]2020Retrospective7,531 imagesDeep CNNCIN2+ detectionSensitivity 91.3%, specificity 80.0%, AUC 0.93
Xue et al [7]2020Retrospective6,763 imagesCNN + attentionCIN2+ detectionSensitivity 98.0%, specificity 62.0%
Zhao et al [8]2022Retrospective3,673 imagesCNNHigh-grade CINSensitivity 86.2%, specificity 78.6%
Liu et al [1]2024Meta-analysis33 studiesMultiple modelsCIN2+ detectionPooled sensitivity 93.0%, specificity 85.0%