影像科学与光化学2025，Vol.43Issue(6)：63-69,7.DOI:10.7517/issn.1674-0475.2025.06.09
多模态医学影像融合与人工智能技术在宫颈癌AI辅助诊断中的应用效果观察
Application of Multi-modal Medical Image Fusion and AI Technology in Assisted Diagnosis of Cervical Cancer
摘要
Abstract
Objective:To explore the application effects of multi-modal medical image fusion and artificial intelligence(AI)assisted diagnostic technology in cervical cancer screening,aiming to improve the detection rate of cervical cancer and precancerous lesions(i.e.sensitivity).Methods:This study included 442 female patients requiring cervical biopsy,employing colposcopy combined with optical coherence tomography(OCT)technology for cervical tissue imaging.Firstly,colposcopy was used to comprehensively observe the cervix and record any abnormal areas.Subsequently,OCT technology was employed to perform point-by-point scanning of the cervix,with real-time analysis of OCT images.Suspicious lesion sites were identified based on colposcopy results and targeted biopsies were performed.Using pathological diagnosis as the gold standard,using colposcopy and OCT examination as routine controls,the positive detection rates of colposcopy alone and combined colposcopy+OCT+AI were compared.The primary evaluation indicators were the detection rates of low-grade squamous intraepithelial lesions(LSIL),high-grade squamous intraepithelial lesions(HSIL),and cervical cancer.Results:The combined method of colposcopy+OCT+AI achieved a detection rate of 92.76%for LSIL lesions,which was an increase of 14.44%compared to colposcopy alone(P<0.05).For HSIL and cervical cancer,the new method reduced the missed diagnosis rates from 16.33%and 10.71%to 4.08%and 1.79%,respectively(P<0.05).Conclusion:Multi-modal medical image fusion combined with AI-assisted diagnostic technology significantly improved the detection rates of cervical cancer and precancerous lesions while reducing the missed diagnosis rates.This approach has the potential to become a powerful tool for improving cervical cancer screening outcomes.Future studies should further expand the sample size to validate the application effects of this technology in different regions and populations.关键词
宫颈癌筛查/多模态医学影像融合/光学相干断层扫描/阴道镜检查Key words
cervical cancer screening/multi-modal medical image fusion/optical coherence tomography/colposcopy分类
医药卫生引用本文复制引用
冉林,谷新,姜方清,谭雪梅,田勇,彭莉芳..多模态医学影像融合与人工智能技术在宫颈癌AI辅助诊断中的应用效果观察[J].影像科学与光化学,2025,43(6):63-69,7.基金项目
2023年恩施州科技计划项目(D20230068). （D20230068）
