Yong-duk Cho

Medical image compression in teleradiology using low-bit-rate communication networks

Teleradiology is defined as the practice of radiology at a distance. Medical images are acquired from one location and are transmitted to one or more distant sites where they are displayed for a diagnosis. Timely availability of medical images via a variety of communication networks is one of the primary goals of teleradiology. In this paper, we propose a medical image compression that can be effectively utilized in the teleradiology system using low-bit rate communication networks. For this purpose, we make use of regions of interest (ROIs) that may be clinically important in medical images. Our study shows that our proposed compression method can reduce the transmission time significantly if the ratio of ROI in the image is small. For example, if the twenty percent of an image belongs to ROI, (ROI ratio equals 0.2), the compression ratio is increased by the scale of about three compared with lossless compression. Accordingly the transmission time is reduced by the same scale. As well, by preserving the clinically important regions, the risk of wrong diagnosis is much reduced compared with lossy compression.

Compression of medical images with regions of interest (ROIs)

In most medical images, regions of interest (ROIs) that may include clinically important information exist and occupy a small portion of the image. Based on this observation, we present compression methods that can effectively compress medical images with ROIs. They are implemented in a manner that ROIs are reversibly compressed and non-ROI (the region outside of ROIs) is irreversibly compressed. In this paper, we present and analyze the three different compression schemes: a DCT-based compression, a DCT/HINT compression, and a HINT-based compression. These methods compress ROIs by reversible compression and non-ROI by irreversible compression. Our current study shows that compression ratio decreases exponentially as ROI ratio (the portion of ROIs in the image) increases. Also, it showed that RMSE (Root-Mean-Squared Error) is not much dependent upon the ROI ratio. To verify this, we tested seven heart X-ray images, twelve head MR images, ten abdomen CT images, and ten chest CT images. Our experimental results showed that the DCT-based compression is the best among the three proposed methods in terms of compression ratio, algorithm complexity, and quality of a reconstructed image.