Aimed to provide an efficient architecture for the storage and processing of medical images, a cloud computing-based system is proposed in this paper. In addition, with the highly increased computational ability of embedded systems, an ARM-based Cortex-A8 embedded platform is applied in the proposed architecture as the kernel for the lossless compression, decompression, as well as the processing of medical images. All the images generated are first sent to the embedded platform for lossless compression before stored in the image database server. On the other hand, images are retrieved and first sent to the platform for decompression, and then transmitted via the network to the client for viewing purpose. In the proposed system, the decompressed images are transmitted with the protocol of HTTP, which means only a browser is needed for the client. Moreover, the client can perform image processing, e.g, image sharpening and histogram equalization with respect to the selected image in the browser, no other application program is required. As no other application program is required for the client other than the browser, the proposed system architecture is very easy to maintain with, and the computational burden traditionally imposed on the client can be quite alleviated. Furthermore, the coding algorithm used is very effective and efficient for the lossless compression of images as we will see in the experiment. The cloud computing-based architecture in conjunction with the high efficient coding algorithm make the proposed system very feasible for practical usage.