Scholarlink

Objectives <p>Delineating the Couinaud segments is a critical component of liver surgery and monitoring that has traditionally relied on labor-intensive methods that are prone to variability. While fully or semi-automatic methods exist, they generally lack accuracy or require extensive post-processing or corrections to the outputs.</p> Methods <p>We present a framework that integrates deep learning-based segmentation with auxiliary landmark identification to create a personalized pictorial model on which to base precise Couinaud landmark localization.&#xa0;Data from 225 non-contrast T1-weighted MRIs from 4 different studies were used to evaluate the performance against benchmark techniques and human-defined ground truth.</p> Results <p>The personalized model outperformed the benchmark method in every landmark placement and Couinaud segment volume estimation, being significantly better in 5/8 landmarks and 7/8 segments.</p> Conclusion <p>The proposed system is explainable, agnostic to imaging modality and is able to incorporate new data without retraining, enhancing its robustness and scalability across diverse clinical contexts. These findings underscore the potential of our framework to substantially improve Couinaud accuracy and streamline clinical workflows, optimizing liver surgery planning and monitoring.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automatic Couinaud segmentation using AI and pictorial representation landmarking

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月30日

查看全文
  • Luis Miguel Núñez,
  • Paul Aljabar,
  • Sir Michael Brady

摘要

Objectives

Delineating the Couinaud segments is a critical component of liver surgery and monitoring that has traditionally relied on labor-intensive methods that are prone to variability. While fully or semi-automatic methods exist, they generally lack accuracy or require extensive post-processing or corrections to the outputs.

Methods

We present a framework that integrates deep learning-based segmentation with auxiliary landmark identification to create a personalized pictorial model on which to base precise Couinaud landmark localization. Data from 225 non-contrast T1-weighted MRIs from 4 different studies were used to evaluate the performance against benchmark techniques and human-defined ground truth.

Results

The personalized model outperformed the benchmark method in every landmark placement and Couinaud segment volume estimation, being significantly better in 5/8 landmarks and 7/8 segments.

Conclusion

The proposed system is explainable, agnostic to imaging modality and is able to incorporate new data without retraining, enhancing its robustness and scalability across diverse clinical contexts. These findings underscore the potential of our framework to substantially improve Couinaud accuracy and streamline clinical workflows, optimizing liver surgery planning and monitoring.