Nomogram based on neutrophil-to-lymphocyte ratio and MR diffusion parameters for predicting Ki67 in hepatocellular carcinoma

Motivation: Preoperative Ki67 prediction in hepatocellular carcinoma (HCC) helps assess tumor aggressiveness and guide treatment. Goal(s): To develop and validate a multiparameter prediction model for Ki67 expression in HCC. Approach: Clinical, pathologic, and MRI data, including intravoxel incoherent motion and diffusion kurtosis imaging parameters, were analyzed for training and validation sets. Logistic regression was used for feature selection, and model evaluation included area under the curve, decision curve analysis, and calibration. Results: The combined model, including neutrophil-to-lymphocyte ratio, tumor margin, pseudocapsule, and MRI parameters, achieved AUC 0.905 (training) and 0.847 (validation), with NLR and ADCslow_Aver as key predictors.. Impact: This study presents a noninvasive model for predicting Ki67 in hepatocellular carcinoma, combining NLR, imaging features, and MRI diffusion parameters to improve preoperative risk stratification. It may reduce biopsy needs, support personalized treatment planning, and aid in monitoring treatment response.