Fuzzy Bit-Plane-Dependence Region Competition

This paper presents a novel variational model based on fuzzy region competition and statistical image variation modeling for image segmentation. In the energy functional of the proposed model, each region is characterized by the pixel-level color feature and region-level spatial/frequency information extracted from various image domains, which are modeled by the windowed bit-plane-dependence probability models. To efficiently minimize the energy functional, we apply an alternating minimization procedure with the use of Chambolle’s fast duality projection algorithm, where the closed-form solutions of the energy functional are obtained. Our method gives soft segmentation result via the fuzzy membership function, and moreover, the use of multi-domain statistical region characterization provides additional information that can enhance the segmentation accuracy. Experimental results indicate that the proposed method has a superior performance and outperforms the current state-of-the-art superpixel-based and deep-learning-based approaches.

MO1045DEVELOPMENT OF A HOME CARE SUPPORT SYSTEM FOR PERITONEAL DIALYSIS PATIENTS I - USING DIGITAL WATERMARKING AND STEGANOGRAPHY*

Background and Aims As it is necessary to refrain from going out due to the COVID-19 pandemic, a system that allows dialysis patients to be treated at a remote location or at home, i.e., a home care support system, is required. Information and communications technology (ICT) used for these purposes is widely applied in various medical fields. Using ICT has the advantage of allowing the sharing of patients’ electronic patient records (EPR) among medical staff, but increases the risk of copyright infringement and privacy leaks during archiving and transmission. We have developed a home care support system for peritoneal dialysis patients using information hiding technology consisting of both digital watermarking technology for copyright protection and steganography technology for communication security when treating patients at home using ICT. In addition, we evaluated the developed system. Method The system for sharing medical information was developed in the PHP programming language on a personal computer system using Microsoft Azure cloud services. Figure 1 shows an explanation of the digital watermarking technology and steganography technology used in the developed system. 1. Digital watermarking technology The patient’s data, such as EPR data, facility name, etc., were hidden in the region of non-interest (RONI) of the patient’s chest CT image series and stored in a database. 2. Steganography technology We call scene photos “cover pictures.” Medical information (CT images, etc.) was hidden in the cover picture. In this study, the cover picture containing the medical information was designated as a Stego image. A body CT image series (16-bit, 512 × 512, 100 slices) was used to verify the steganography technique. These CT images were compressed using 7-Zip and then saved in a folder, which was then embedded in the cover photo. The Stego image was then sent from the patient’s home to the medical institution via the home care support system. Results We investigated the hash value, peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) of the image series (Fig. 2). If the structure of the image or photograph was exactly the same, the SSIM shows a value of 1. If the PSNR is ≥ 40 dB, the image quality can be maintained without affecting diagnosis. If part of the ROI is changed during transmission, the hash value decoded from the received Stego image will be different from that before transmission. For Stego images containing watermarked or hidden CT images with 4000 words embedded, SSIM and PSNR were ≥ 0.99 dB and 65.3 dB, respectively. If the medical information was embedded in a low bit plane, such as a 1-bit or 2-bit plane, the radiologist could not identify the embedded information. When our technology was applied, there were no changes in the capacity of CT images or Stego images before and after embedding. Therefore, it was not possible to tell that medical information was embedded due to changes in capacity. Conclusion Using ICT, we have built a home care support system that can conceal medical information by combining digital watermarking technology and steganography technology to ensure the copyright of images and to ensure privacy and secure transmission of EPR and CT images. Using the developed system, daily medical information of dialysis patients could be transmitted safely to the institute, and the medical staff could share the information safely. Both techniques can be applied to all digital image information, and is not just limited to CT images.

Progressive Transmission of Medical Images via a Bank of Generative Adversarial Networks

The healthcare sector is currently undergoing a major transformation due to the recent advances in deep learning and artificial intelligence. Despite a significant breakthrough in medical imaging and diagnosis, there are still many open issues and undeveloped applications in the healthcare domain. In particular, transmission of a large volume of medical images proves to be a challenging and time-consuming problem, and yet no prior studies have investigated the use of deep neural networks towards this task. The purpose of this paper is to introduce and develop a deep-learning approach for the efficient transmission of medical images, with a particular interest in the progressive coding of bit-planes. We establish a connection between bit-plane synthesis and image-to-image translation and propose a two-step pipeline for progressive image transmission. First, a bank of generative adversarial networks is trained for predicting bit-planes in a top-down manner, and then prediction residuals are encoded with a tailored adaptive lossless compression algorithm. Experimental results validate the effectiveness of the network bank for generating an accurate low-order bit-plane from high-order bit-planes and demonstrate an advantage of the tailored compression algorithm over conventional arithmetic coding for this special type of prediction residuals in terms of compression ratio.