目的 探讨近红外光谱脑功能成像技术在评估稳定期精神分裂症患者执行言语流畅性任务时前额与双侧颞区的脑功能特征，及其在辅助诊断精神分裂症的临床应用。 方法 纳入45例稳定期精神分裂症患者及30例健康志愿者。入组时应用近红外光谱脑功能成像检测执行言语流畅性任务时的额叶和双侧颞叶脑激活特征，测定额叶和双侧颞叶积分值、重心值及斜率。比较患者组与健康对照组在一般资料、组词数及近红外光谱脑功能成像指标的差异，应用logistic回归模型及受试者工作特征（ROC）曲线分析近红外光谱脑功能成像指标对精神分裂症患病的影响及诊断效果。 结果 在执行言语流畅性任务时，患者组组词数低于健康对照组，其额叶积分值、斜率和双侧颞叶积分值均低于对照组，差异均具有统计学意义（均P < 0.05）；而额叶重心值、颞叶重心值和双侧颞叶斜率在两组间差异无统计学意义（均P > 0.05）。二元logistic回归模型表明颞叶积分值与精神分裂症患病相关（OR = 0.979， P = 0.005）；ROC曲线分析显示额叶积分值、额叶斜率和颞叶积分值对是否患病分类较好，曲线下面积（AUC）分别为0.681、0.644和0.779（均P < 0.05）。 结论 稳定期精神分裂症在执行言语流畅性任务时额叶及颞叶的积分值降低、额叶斜率降低；这些指标对区分精神分裂症患者与健康对照者有一定的临床价值。

Objective To investigate the functional brain characteristics of the frontal and bilateral temporal regions during a verbal fluency task （VFT） in patients with stable schizophrenia using functional near-infrared spectroscopy （fNIRS）， and to evaluate its potential clinical application in assisting the diagnosis of schizophrenia. Methods A total of 45 stable schizophrenia patients and 30 healthy controls were enrolled in the study. fNIRS was employed to assess brain activation patterns in the frontal and bilateral temporal lobes during the VFT. The integral values， centroid values， and slopes of the frontal and bilateral temporal lobes were measured. Differences in demographic characteristics， word count， and fNIRS-derived parameters between the two groups were analyzed. A logistic regression model and receiver operating characteristic （ROC） curve analysis were conducted to evaluate the diagnostic performance of the fNIRS measures in distinguishing schizophrenia patients from healthy controls. Results During VFT， the patient group demonstrated reduced word count and decreased integral values and slopes in the frontal lobe， as well as lower integral values in the bilateral temporal lobes， compared to the healthy control group （all P < 0.05）. However， no statistically significant differences were observed between the two groups in terms of frontal lobe centroid value， temporal lobe centroid value， or bilateral temporal lobe slope （all P > 0.05）. The logistic regression model indicated that the temporal lobe integral value was significantly associated with the diagnosis of schizophrenia （OR = 0.979， P = 0.005）. ROC curve analysis revealed that the frontal lobe integral value， frontal lobe slope， and temporal lobe integral value exhibited acceptable discriminatory power for disease classification， with area under the curve （AUC） values of 0.681， 0.644， and 0.779， respectively （all P < 0.05）. Conclusion In stable schizophrenia， the integral values of the frontal and temporal lobes， as well as the slope of the frontal lobe， decrease during the VFT， suggesting potential clinical utility in differentiating individuals with schizophrenia from healthy controls.