Brain slice representation via feature point maps and retrieval using similarity distance metrics
Aman Kumar Srivastava, Priyanka Gupta, Parthasarathi Mangipudi · 2016
In this paper, we propose a medical image retrieval scheme which could be of immense help to the medical practitioners. Large collection of medical images are derived from various sources like X-rays, CT scan, PET scan, ultrasound, MRI etc and are available in different contexts. The real challenge then is the accessibility to such a database and retrieval of the relevant images in a quick time which would help the medical practitioners and experts for better diagnosis. The basic idea in the proposed framework is to input a 2D Brain Magnetic Resonance (MR) query slice and output visually similar slices pertaining to the MR image database. The edge information of the lesions is obtained by combining multi-resolution methods and spatial methods. Wavelet methods are known to be robust to bias fields present in magnetic resonance. The local structure representation is then utilized for retrieval of similar slices from the brain volumes. Hausdorff and chamfer distances were used as similarity distances metrics for retrieval.