Related Experiment Video
Updated: Jul 12, 2026
Chromogenic In Situ Hybridization as a Tool for HPV-Related Head and Neck Cancer Diagnosis
Published on: June 14, 2019
Artificial Intelligence for Cervical Human Papillomavirus Infection and Lesion Screening: A Cross-Sectional Analysis
Ping Zheng1, Yan Yang2,3, Sha Li2
1Department of Obstetrics and Gynecology, Beijing Chao-Yang Hospital, Capital Medical University, Beijing, China, 915979411@qq.com.
Objective:
This cross-sectional study aimed to explore the application potential of artificial intelligence (AI) in screening and diagnosing cervical human papillomavirus (HPV) infection and lesions, and to assess patient satisfaction with the current diagnostic and therapeutic process as well as their unmet needs.
Design:
This was a cross-sectional observational study. Participants/Materials: A total of 308 valid female participants who received cervical HPV and lesion screening were enrolled. Research material consisted of self-designed questionnaires (online and paper versions) covering demographic information, screening experience, cognition and acceptance of AI-assisted cervical screening, and open-ended feedback.
Setting:
Questionnaires were distributed via online Questionnaire Star platform and offline paper forms.
Methods:
An online cross-sectional survey was conducted via the Questionnaire Star platform, and 308 valid responses were collected. Descriptive statistics were used to summarize participants' demographic characteristics and questionnaire responses. Chi-square tests were performed to examine associations between demographic factors (e.g., age, residence) and key outcomes (e.g., AI acceptance, primary concerns). A two-tailed p value <0.05 was considered statistically significant. Qualitative content analysis was applied to synthesize and interpret responses to open-ended questions.
Results:
Most respondents were women aged 25-35 years (34.74%), with 84.74% residing in urban areas. Among all participants, 76.30% reported a history of HPV infection, and 91.56% had undergone ThinPrep cytologic test (TCT). The most distressing part of the screening process was anxiety during result waiting (41.23%), and 58.12% found medical terminology difficult to understand. Although 61.04% of respondents had no prior knowledge of AI-assisted diagnosis, 58.77% were willing to learn about its application in improving diagnostic efficiency. Younger respondents (≤35 years) showed significantly higher willingness to learn about AI than those aged >35 years (65.1% vs. 52.4%, χ2 = 6.24, p = 0.012). Additionally, 75.97% of respondents believed AI could shorten result waiting times, and 73.70% trusted the "AI preliminary screening + physician confirmation" model. The top concerns regarding AI application were technical reliability (70.78%) and data privacy (68.18%).
Limitations:
This study has several limitations, including convenience sampling of urban, HPV-prone women, lack of standardized AI knowledge assessment and demonstration, reliance on self-reported screening history, a cross-sectional design that precludes causal inference, and a modest sample size (n = 308) for subgroup analyses. However, these issues are common in exploratory research and provide a clear roadmap for future work: larger and more representative samples, baseline knowledge tests, medical record verification, and longitudinal designs to validate and extend our findings.
Conclusions:
Patients with cervical lesions have strong demands for diagnostic efficiency, psychological support, and information transparency. AI technology holds great potential in enhancing screening efficiency and assisting diagnosis; however, key challenges remain, including ensuring data privacy, improving technical reliability, and strengthening patient trust.