In　regard　of　problems　of　medical　care　big　data　such　as　dispersed　data,　weak　consistency,　low　electronic　degree　and　low　visualization　degree,　a　set　of　innovative　image　recognition,　data　collection,　arrangement　and　storage　projects　were　put　forward.　Data　collection　method,　which　adopts　OCR　recognition　algorithm　on　the　basis　of　CRNN,　together　with　human-assisted　man-machine　recognition　method,　realized　amazing　recognition　accuracy　and　had　the　ability　to　update　and　continuously　and　improve　accuracy.　Through　practical　application　on　parts　of　hospitals　in　Shenzhen,　the　system　makes　it　possible　for　patients　to　realize　electronization　and　datamation　of　heath　records,　improve　patients’　understanding　to　their　heath,　and　guide　doctors　to　diagnose　health　of　patients.　Moreover,　continuous　collection　and　arrangement　of　data　of　health　records　can　provide　effective　methods　and　data　basis　to　accumulation　and　discovery　of　big　data　of　the　medical　industry.　©　2018　Association　for　Computing　Machinery.