Abstract: Objective To explore a method for early detection of diabetic autonomic neuropathy (DAN) on the basis of analysis of galvanic skin response (GSR).Methods In 25 randomly selected diabetic inpatients (14 with clinical DAN symptoms),GSR signals were collected from the fingers of both hands based on Virtual Instrumental WorkBench-LabVIEW and compared with those acquired from normal subjects. Results The GSR indices of patients either with or without DAN were different from those of normal subjects,and even in the same patients,the changes of these indices after clinical treatment were also obvious.Conclusion GSR detection and analysis can help understand the impairment of the autonomic system and facilitate DAN assessment.