Objective To explore the role of the E-mental health service platform "mood thermometer" in the recognition, diagnosis and treatment of depression disorders in community setting. Methods Mood thermometer application (App) is an online autonomous screening and auxiliary diagnosis system. A total of 700 outpatients from 2 community health service centers in Hongkou District of Shanghai were evaluated with the self-designed questionnaire, patient health questionnaire-9 (PHQ) and the mini-international neuropsychiatric interview (MINI), which had been set in the App for detection of depression disorder. Subsequent offline diagnosis and treatment of the patients were followed up 3 months later. Results The recognition rate of depressive symptoms was 14.86% (104/700) as detected by PHQ-9, and the detection rate of depressive disorder was 7.86% (55/700) as detected by MINI. Among the 55 patients with depression, 46 had unipolar depression, 8 had bipolar depression, and 1 had dysthymia. Of the 46 patients with unipolar depression, 8 were previously diagnosed, and 1 of the 8 patients with bipolar depression was previously diagnosed with depression, with a clinical diagnosis rate of 1.29% (9/700). The detection rate of MINI for depressive disorder was 6.57% higher than the clinical diagnosis rate. The treatment rate of patients with depression disorder was 16.36% (9/55) at baseline and increased by 10.91% (6/55) 3 months later, with the diagnosis and treatment rate increased to 27.27% (15/55). Conclusion The E-mental health platform in the community screening and diagnosis evaluation can not only improve the detection rate of depression disorder, but also promote the subsequent actual clinical diagnosis and treatment rate. E-mental health assessment should be integrated in the primary health care services to promote early diagnosis and effective treatment of depression disorder.