Objective To construct a prognostic prediction model for patients with facial nerve injury based on neu-roelectrophysiological parameters,providing an objective reference for early clinical prognosis assessment and personalized treatment planning.Methods A total of 146 patients with facial nerve injury admitted to the Central Hospital Affiliated to Dalian University of Technology(Dalian Central Hospital)from January 2022 to November 2024 were enrolled.All patients received standardized treatment and eye care.Clinical characteristics were assessed before treatment.Electroneurography(ENoG),blink reflex(BR),and surface electromyography(sEMG)were performed to obtain key neuroelectrophysiological in-dicators,including compound muscle action potential(CMAP)amplitude and latency,R1 wave latency and amplitude,F-wave latency and occurrence rate,resting root mean square(RMS)value,and integrated electromyography(iEMG)during voluntary contraction.At six months after treatment,the House-Brackmann(HB)grading system was used to evaluate prognosis.Pearson or Spearman correlation analysis was applied to examine the correlations between neuroelectrophysiological indicators and baseline HB grade Ⅳ–Ⅵ as well as baseline Sunnybrook score.Multivariate logistic regression analysis was used to identify risk factors for poor prognosis.Based on the regression results,a nomogram prediction model was constructed.Inter-nal validation was performed using the bootstrap method(resampling times=1,000).The discrimination,calibration,and clin-ical utility of the model were assessed by calibration curve,receiver operating characteristic(ROC)curve,and decision curve analysis(DCA).Results At 6 months after onset,116 patients(79.45%)had a good prognosis and 30(20.55%)had a poor prognosis.Compared with the good prognosis group,the poor prognosis group had significantly longer CMAP latency,R1 wave latency,and F-wave latency,higher resting RMS,and significantly lower CMAP amplitude,R1 wave amplitude,F-wave occurrence rate,and iEMG during contraction(t/P=13.539/<0.001,13.464/<0.001,4.659/<0.001,14.476/<0.001,12.431/<0.001,9.604/<0.001,14.931/<0.001,7.567/<0.001).CMAP amplitude,R1 wave amplitude,F-wave occurrence rate,and iEMG during contraction were significantly negatively correlated with baseline HB grade(rs=-0.860 to-0.675,all P<0.001)and significantly positively correlated with baseline Sunnybrook score(r=0.643 to 0.832,all P<0.001).CMAP latency,R1 wave latency,F-wave latency,and resting RMS were significantly positively correlated with baseline HB grade(rs=0.516 to 0.847,all P<0.001)and significantly negatively correlated with baseline Sunnybrook score(r=-0.815 to-0.478,all P<0.001).Logis-tic regression analysis showed that CMAP latency,R1 wave latency,and resting RMS were independent risk factors for poor prognosis in patients with facial nerve injury(all P<0.05).Conclusion The nomogram prediction model based on CMAP latency,R1 wave latency,and resting RMS demonstrates excellent predictive performance and strong clinical utility for prog-nostic assessment in patients with facial nerve injury.It provides a scientific basis for early identification of high-risk patients and optimization of treatment strategies.