What question did this study set out to answer?

The study aims to enhance risk stratification for patients undergoing resection of colorectal liver metastases using multimodal prediction models.

July 16, 2026Scientific ReportsOpen Access

Improving preoperative risk stratification in colorectal liver metastases: a multi-institutional evaluation of multimodal prediction models

View Full Paper

Ask AI

Bookmark

Authors

Retrospective cohort study evaluates multimodal prediction models for liver survival outcomes, suggesting improved risk stratification methods.

The study aims to enhance risk stratification for patients undergoing resection of colorectal liver metastases using multimodal prediction models.
Conducted a retrospective cohort study with 1,301 consecutive patients from two cancer centers.
Incorporated clinical and computed tomography (CT) data into radiomic and deep learning models.
Compared the performance of unimodal and multimodal prediction models in predicting hepatic disease-free survival and overall survival.
Deep learning model utilizing CT imaging achieved a C-index of 0.61 for both hepatic disease-free survival and overall survival.
The multimodal ‘clinicoradiomic’ model outperformed unimodal models with a C-index of 0.63 compared to 0.61 and the current clinical risk score at 0.55.
Preoperative imaging models effectively predict risk of hepatic recurrence, surpassing all existing models.

Kobayashi et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Retrospective cohort study evaluates multimodal prediction models for liver survival outcomes, suggesting improved risk stratification methods.