To develop and validate a CT-based radiomics nomogram for preoperative prediction of lymphovascular invasion (LVI) in colorectal cancer (CRC), highlighting the importance of LVI in treatment decisions.
Key Findings:
27.78% of patients in the training set and 28.70% in the validation set were LVI-positive.
Independent predictors of LVI included tumor volume, maximum tumor diameter, depth of invasion, lymph node diameter, CEA level, NLR, standard deviation of CT values, and Rad-score.
The nomogram showed good predictive performance with AUCs of 0.776 (95% CI: X-Y) in training and 0.722 (95% CI: A-B) in validation sets.
Interpretation:
The CT-based radiomics nomogram provides a reliable non-invasive tool for predicting LVI in CRC, aiding in personalized treatment strategies by identifying high-risk patients.
Limitations:
Retrospective design may introduce selection bias.
Single-center study limits generalizability and may affect the applicability of findings to broader populations.
Further validation in diverse populations is needed.
Conclusion:
The proposed nomogram enhances preoperative LVI prediction in CRC, potentially improving individualized clinical strategies and prognosis assessment.
