Articles In Press
Back to the articles list |
Back to browse issues page
Ethics code: مقاله مروری است و کد اخلاق ندارد
Shahroud University of Medical Sciences
Abstract: (26 Views)
Objective: This study aimed to evaluate the capabilities and challenges of image processing in telemedicine.
Information Sources: A literature search was conducted in PubMed, Scopus, and Web of Science, and 38 articles were selected based on the predefined inclusion criteria.
Study Selection: Data were extracted across five main domains: (1) types of image processing algorithms, (2) clinical evaluation, (3) physician involvement in the research team, and (4) technical and economic requirements of the systems.
Results: The findings of this systematic review indicated that among the 38 included studies, the most frequently used approaches were deep learning algorithms, particularly convolutional neural networks (CNNs). Other machine learning models such as SVM and hybrid approaches were used less frequently. Regarding learning paradigms, most studies employed EAGER algorithms, whereas only a small number reported LAZY approaches. In addition, only 18% of the studies performed external validation, and 39% reported physician participation.
Conclusion: The results indicate that none of the included studies reported a cost analysis or provided a comprehensive description of the technical infrastructure, representing a major gap in the existing literature. Although image processing in telemedicine has achieved substantial algorithmic performance improvements, limited interpretability, restricted clinical evaluation, lack of economic assessment, and insufficient reporting of technical requirements remain key barriers to large-scale real-world implementation. Therefore, future research should focus on more comprehensive evaluations and cost–benefit analyses to support the development of reliable and cost-effective telemedicine image processing systems.
Information Sources: A literature search was conducted in PubMed, Scopus, and Web of Science, and 38 articles were selected based on the predefined inclusion criteria.
Study Selection: Data were extracted across five main domains: (1) types of image processing algorithms, (2) clinical evaluation, (3) physician involvement in the research team, and (4) technical and economic requirements of the systems.
Results: The findings of this systematic review indicated that among the 38 included studies, the most frequently used approaches were deep learning algorithms, particularly convolutional neural networks (CNNs). Other machine learning models such as SVM and hybrid approaches were used less frequently. Regarding learning paradigms, most studies employed EAGER algorithms, whereas only a small number reported LAZY approaches. In addition, only 18% of the studies performed external validation, and 39% reported physician participation.
Conclusion: The results indicate that none of the included studies reported a cost analysis or provided a comprehensive description of the technical infrastructure, representing a major gap in the existing literature. Although image processing in telemedicine has achieved substantial algorithmic performance improvements, limited interpretability, restricted clinical evaluation, lack of economic assessment, and insufficient reporting of technical requirements remain key barriers to large-scale real-world implementation. Therefore, future research should focus on more comprehensive evaluations and cost–benefit analyses to support the development of reliable and cost-effective telemedicine image processing systems.
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
Contact Us
Post Address: Bandar Abbas, Imam Hossein Boulevard, campus of University of Medical Sciences, School of Nursing and Midwifery.
Tel: +(98) 763 331 7723
Cell Phone: +(98) 939 654 9816
jmis
hums.ac.ir
