Long noncoding RNAs (lncRNAs) may serve as potential molecular diagnostic markers to improve the capacity of earlier and more accurate diagnosis of dilated cardiomyopathy (DCM). We integrated five independent transcriptomic datasets ( = 504) from Gene Expression Omnibus for systematic identification of lncRNA-based diagnostic biomarkers in DCM. The multivariate logistic regression model based on the six lncRNAs (, , , , , and ) in the ceRNA networks showed high sensitivity and specificity (area under curves >0.8,  < 0.0001) of DCM diagnosis in the training and validation datasets. Functional analysis revealed that the autophagy, protein acetyltransferase, and DNA polymerase activity were associated with high levels of the six-lncRNA signature, while the collagen trimer, extracellular matrix structural constituent, and MHC protein complex were associated with low levels of the signature. Pathway analysis showed that high levels of the six-lncRNA signature were associated with upregulated selective autophagy, interleukin 17 signalings, and extracellular matrix interactions, while were associated with downregulated extracellular matrix organization and collagen formation. The identified six-lncRNA signature, with high performance in molecular diagnosis of DCM, might be applied in future clinical practices combined with traditional markers.