In　the　medicine　practice,　due　to　the　privacy　and　safety　of　electronic　medical　record　(EMR),　the　sharing,　research　and　application　of　EMR　have　been　hindered　to　a　certain　extent.　Thus,　it　becomes　increasingly　important　to　study　semantic　electronic　medical　data　integration,　so　as　to　meet　the　needs　of　doctors　and　researchers　and　help　them　quickly　access　high-quality　information.　This　paper　focuses　on　the　realization　of　semantic　EMRs.　It　shows　how　to　uses　APDG　(Advanced　Patient　Data　Generator)　to　create　a　set　of　virtual　patient　data　for　depression.　Furthermore,　it　explains　how　to　develop　clinical　and　semantic　description　rules　to　construct　semantic　EMRs　for　depression　and　discusses　how　those　generated　virtual　patient　data　can　be　used　in　the　system　of　Smart　Ward　for　the　test　and　demonstration,　without　violating　the　legal　issues　(e.g.,　privacy　and　security)　of　patient　data.