Detecting Cervical Lesions with AI in Colposcopic Images
Author Information
Author(s): Wang Li, Chen Ruiyun, Weng Jingjing, Li Huiping, Ying Shi, Zhang Jinghui, Yu Zehao, Peng Chengbin, Zheng Siming
Primary Institution: Ningbo Medical Centre Lihuili Hospital
Hypothesis
This study aims to investigate the feasibility of employing artificial intelligence models for the detection and localization of cervical lesions by leveraging deep semantic features extracted from colposcopic images.
Conclusion
The proposed approach demonstrates promising capabilities in identifying normal and cervical lesions, particularly excelling in lesion area segmentation.
Supporting Evidence
- The model achieved high sensitivity (96.38%), specificity (95.84%), precision (97.56%), and F1 score (96.96%).
- The model demonstrated an impressive accuracy of 96.18% in distinguishing between normal and lesion images.
- Experimental findings showed superior accuracy compared to algorithms solely based on image segmentation or classification.
- The model accurately estimated lesion areas, providing valuable guidance to assisting physicians.
Takeaway
This study used AI to help doctors find and understand cervical lesions in images, making it easier to decide on treatment.
Methodology
The study employed a segmentation-based deep learning architecture, utilizing a deep decoding network to integrate prior features and establish a semantic segmentation model capable of distinguishing normal and pathological changes.
Potential Biases
The study mitigated potential bias by using images from two different hospitals.
Limitations
The generalizability of the findings may be constrained despite efforts to diversify sources.
Participant Demographics
Images were collected from women undergoing colposcopy at two hospitals in Ningbo, China.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website