Objective To investigate the predictive value of preoperative magnetic resonance imaging (MRI) and clinical characteristics for lymph node metastasis in patients with ovarian cancer.Methods The clinical data of 102 ovarian cancer patients who received treatment at our hospital from January 2022 to January 2025 were retrospectively analyzed. All patients underwent cytoreductive surgery, and pathological specimens were examined. Patients with lymph node metastasis confirmed by pathology were assigned to the observation group (n = 39), and those without lymph node metastasis served as the control group (n = 63). Preoperative MRI findings and clinicopathological data were compared between the two groups. Risk factors for lymph node metastasis were identified by multivariate logistic regression analysis, and a nomogram prediction model was established. The performance of the model was evaluated using calibration curves and ROC curves.Results The proportion of patients in the observation group with stage III-IV, bilateral lesions, low differentiation, and postoperative residual lesion diameter > 2 cm was higher than that in the control group (all P < 0.05). Multivariate logistic regression analysis showed that stage III-IV [O^R = 4.622 (95% CI: 1.647, 12.970) ] and bilateral lesion location [O^R = 6.182 (95% CI: 1.998, 19.128) ] were independent risk factors for lymph node metastasis in ovarian cancer patients (P < 0.05).Conclusion The prediction model constructed by preoperative MRI combined with clinical features has high predictive value for lymph node metastasis in ovarian cancer patients, and can provide a reference for early clinical evaluation and treatment decision-making.