Research on Medical Image Retrieval Based on Arbitrary Shape

Nowadays, with the rapid development of medical imaging technology and information technology more and more medical images are available. Medical image retrieval plays increasingly important role in medical applications. Content-based medical image retrieval (CBMIR) poses unique challenges due to the unique characteristics of medical images. While current research of CBMIR focuses on low-level visual features extraction, we propose a novel medical image retrieval algorithm based on arbitrary shape by combining low-level visual features extraction and mirror-padding, which has the merit of relieving the spurious high frequency components as introduced by no-padding. The efficiency of the proposed algorithm for medical image retrieval is analyzed and verified through experimental results.