A Proposed Deep-Learning-Based Framework for Medical Image Communication, Storage and Diagnosis

• Published in: Wireless personal communications Vol. 131; no. 4; pp. 2331 - 2369
• Main Authors: Ebied, Mostafa, Elmisery, F. A., El-Hag, Noha A., Sedik, Ahmed, El-Shafai, Walid, El-Banby, Ghada M., Soltan, Eman, Al-Zubi, Nayel, Brisha, Ayman, Zekry, Abdelhalim, Khalaf, Ashraf A. M., El-Fishawy, Adel S., El-Rabaie, El-Sayed M., Al-Nuaimy, Waleed, El-Samie, Fathi E. Abd
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.08.2023; Springer Nature B.V
• Subjects: Algorithms; Communications Engineering; Computer Communication Networks; Deep learning; Developing countries; Diagnosis; Engineering; Image compression; Image quality; Image reconstruction; Image storage; Interpolation; LDCs; Medical imaging; Networks; Signal,Image and Speech Processing; Storage facilities; Deep learning; Reconstruction; Interpolation; SIFT; Compression; Medical images; Diagnosis; Decimation
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-022-09931-4

Medical images rather than any other types of images need high storage space. The lack of storage facilities, especially in developing countries, encourages researchers to find solutions for this problem. Compression of medical images is a priority, although it leads to some loss in the stored images. This paper introduces a framework for medical image storage and retrieval for the purpose of diagnosis. This framework adopts decimation as a tool for image compression, while interpolation is used as a tool for further image reconstruction. The quality of the reconstructed images is evaluated with a scale-invariant feature transform (SIFT)-based technique. Another task involved in this paper is the automatic diagnosis from the reconstructed images based on deep learning. Different types of interpolation algorithms are investigated and compared in this framework for the process of medical image reconstruction.