The　current　data　mining　is　confronted　with　many　problems　in　its　application　in　medical　quality　and　safety　monitoring　based　on　traditional　BP　neural　network.　This　paper　is　aimed　at　proposing　a　scheme　based　on　improved　BP　neural　network　while　providing　an　analysis　of　issues　concerning　data　mining　based　on　improved　BP　neural　network　in　medical　quality　supervision.　For　medical　quality　supervision,　the　use　of　improves　BP　neural　network　mining,　not　only　improves　the　level　of　medical　treatment　regulation,　resulting　in　medical　claims　expenses　decreased　by　8%　and　infectious　disease　infection　rate　decreased　by　12%,　but　exerts　positive　effects　in　application　as　well.　In　general,　in　medical　quality　supervision,　the　application　data　mining　based　on　improved　BP　neural　network　plays　a　positive　role　in　two　aspects:　on　the　one　hand,　it　can　effectively　dig　out　useful　data　and　information;　on　the　other　hand,　it　can　guarantee　the　quality　of　medical　quality　supervision.