Conserved regions, or motifs, present among RNA secondary structures serve as a useful indicator for predicting the functionality of the RNA molecules. Automated detection or discovery of these conserved regions is emerging as an important research topic in health and disease informatics. In this short paper we present a new approach for detecting conserved regions in RNA secondary structures by the use of constrained alignment and apply the approach to finding structural motifs in some viral genomes. Our experimental results show that the proposed approach is capable of efficiently detecting conserved regions in the viral genomes and is comparable to existing methods. We implement our constrained structure alignment algorithm into a web server, called RADAR. This web server is fully operational and accessible on the Internet at http://datalab.njit.edu/biodata/rna/RSmatch/server.htm.