Abstract: Objective: To investigate the value of a computed tomography (CT) radiomics nomogram for predicting disease-free survival (DFS) in patients with stage I–III colorectal cancer (CRC) after radical surgery. Methods: Overall, 324 patients with CRC who underwent radical surgery were retrospectively included and grouped into training and validation cohorts at a 7:3 ratio. The least absolute shrinkage and selection operator Cox regression algorithm was employed to select the relevant CT radiomics features. Using Cox regression analysis, clinically significant risk factors were identified and combined with radiomics features to develop a comprehensive nomogram. The predictive performance of the nomogram was evaluated using the C-index, calibration curves, and decision curves, with DFS probabilities estimated using the Kaplan–Meier method. Results: A clinical model was constructed based on three clinical risk factors: pathological N stage, perineural invasion, and KRAS mutation. We achieved a C-index of 0.709 (95% confidence interval CI: 0.678–0.740) in the training cohort and 0.696 (95% CI: 0.644–0.748) in the validation cohort. A nomogram was subsequently developed using the 15 retained radiomics features along with these three clinical risk factors. The nomogram demonstrated superior predictive performance, with a C-index of 0.820 (95% CI: 0.799–0.841) in the training cohort and 0.818 (95% CI: 0.775–0.861) in the validation cohort, thus outperforming the clinical model. Calibration curves indicated good agreement between the predicted and observed DFS. Decision curve analysis further confirmed the greater net benefit of the nomogram in clinical applications. Conclusion: The constructed integrated nomogram demonstrated high discriminative ability, good calibration, and greater net benefit in the individualized prediction of postoperative DFS in patients with stage I–III CRC cancer. It maintained a robust predictive performance in the validation cohort and outperformed the clinical model.