Cervical‐YOSA: Utilizing prompt engineering and pre‐trained large‐scale models for automated segmentation of multi‐sequence MRI images in cervical cancer

• Published in: IET image processing Vol. 18; no. 12; pp. 3556 - 3569
• Main Authors: Xia, Yanwei, Ou, Zhengjie, Tan, Lihua, Liu, Qiang, Cui, Yanfen, Teng, Da, Zhao, Dan
• Format: Journal Article
• Language: English
• Published: Wiley 01.10.2024
• Subjects: biomedical MRI; image segmentation; transforms
• ISSN: 1751-9659; 1751-9667
• DOI: 10.1049/ipr2.13194

Cervical cancer is a major health concern, particularly in developing countries with limited medical resources. This study introduces two models aimed at improving cervical tumor segmentation: a semi‐automatic model that fine‐tunes the Segment Anything Model (SAM) and a fully automated model designed for efficiency. Evaluations were conducted using a dataset of 8586 magnetic resonance imaging (MRI) slices, where the semi‐automatic model achieved a Dice Similarity Coefficient (DSC) of 0.9097, demonstrating high accuracy. The fully automated model also performed robustly with a DSC of 0.8526, outperforming existing methods. These models offer significant potential to enhance cervical cancer diagnosis and treatment, especially in resource‐limited settings. This article introduces two innovative models that leverage artificial intelligence, including pre‐trained models and feature fusion, to achieve superior accuracy and robustness in the automated segmentation of multi‐sequence cervical cancer images.