An efficient architecture for medical high-resolution images transmission in mobile telemedicine systems

Comput Methods Programs Biomed. 2020 Apr:187:105088. doi: 10.1016/j.cmpb.2019.105088. Epub 2019 Nov 26.

Abstract

Background and objective: The medical high-resolution image is very important in image processing and computer vision applications, which plays a critical role in image-guided diagnosis, clinical trials, consultation, and case discussion. How to efficiently access medical high-resolution images in mobile telemedicine systems is becoming a big challenge. Therefore, this work proposes an efficient pyramid architecture for optimizing medical high-resolution images transmission and rendering.

Methods: The proposed architecture consists of three core schemes: (1) unbalance pyramid scheme based on geometric relationship, (2) indexing scheme based on hash table and lattice partitioning and (3) query scheme based on similar matching. Then, we design the responsive service components: generating service, indexing service, and query service. Finally, these services are combined into a prototype system that enables efficient transmission and rendering of medical high-resolution images.

Results: The result shows that the unbalance pyramid scheme can quickly generate the pyramid structure and the corresponding image files. The indexing scheme can create the index structure and the index file in real-time. The query scheme can not only match the best layer to which the image block belongs in real-time, but also can accurately capture the query image block.

Conclusions: The prototype system based on proposed architecture is fully compliant with the DICOM standard, which can be seamlessly integrated with other existing medical systems or mobile applications, and used in various scenarios such as diagnosis, research, and education.

Keywords: DICOM and DICOM web service; Image transmission; Medical high-resolution image; Mobile telemedicine; Unbalance image pyramid.

MeSH terms

  • Algorithms
  • Computer Communication Networks*
  • Diagnostic Imaging
  • Humans
  • Image Processing, Computer-Assisted / methods*
  • Information Storage and Retrieval
  • Machine Learning
  • Models, Statistical
  • Neural Networks, Computer
  • Radiology Information Systems
  • Remote Consultation / instrumentation
  • Software
  • Telemedicine / instrumentation*
  • Telemedicine / methods*
  • User-Computer Interface*