Infantile hemangioma (IH) is the most common vascular tumor among infants and children. However, the understanding of pathogenesis about IH has not been fully elucidated, and the potential diagnostic maker remains further explored. In this study, we aimed to find miRNAs as potential biomarkers of IH through bioinformatic analysis. The microarray datasets GSE69136, GSE100682 were downloaded from the GEO database. The co-expressed differential miRNAs were identified by analyzing these two datasets. The downstream common target genes were predicted by the ENCORI, Mirgene, miRWalk, and Targetscan databases. GO annotation and KEGG pathway enrichment analysis for target genes were performed. The STRING database and Cytoscape software were used to construct the protein-protein interaction network and screen hub genes. Then potential diagnostic markers for IH were further screened and identified by using Receiver operating characteristic curve analysis. A total of thirteen co-expressed up-regulated miRNAs were screened out in the above two datasets, and 778 down-regulated target genes were then predicted. GO annotation and KEGG pathway enrichment analysis indicated that the common target genes strongly correlated with IH. Through the DEM-hub gene network construction, six miRNAs associated with the hub genes were identified. Finally, has-miR-522-3p, has-miR-512-3p, has-miR-520a-5p with high diagnostic values were screened out by receiver operating characteristic analysis. In the study, the potential miRNA-mRNA regulatory network was firstly constructed in IH. And, the three miRNAs might be used as potential biomarkers for IH, which also provided novel strategies for the therapeutic intervention of IH.