The opportunities and challenges in deploying AI in medical imaging
Zhu, Huayun (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025112329162
https://urn.fi/URN:NBN:fi:amk-2025112329162
Tiivistelmä
In recent years, artificial intelligence (AI) technology has made significant progress in the medical field, especially in the aspect of medical imaging, demonstrating great potential. By analysing medical images through AI algorithms, the AI system can assist doctors in identifying diseases more quickly and accurately. Although the prospects are promising, the application of artificial intelligence in medical imaging still faces many challenges. Issues related to data privacy and security need to be addressed urgently, and the transparency and explainability of algorithms also need to be improved.
This study aims to systematically explore the opportunity arising from the integration of AI in medical imaging and to identify the specific challenges healthcare organizations face in its deployment as well. The purpose of this study is to provide a synthesized understanding of how AI can be effectively and ethically utilized in medical imaging by integrating existing research data.
An umbrella review methodology was employed to collect and analyse evidence from 15 systematic review published between 2019 and 2024. Articles were collected from PubMed and ScienceDirect using defined inclusion and exclusion criteria. Data extraction and content analysis were conducted to identify recurring themes related to AI opportunities and challenges. The Critical Appraisal Skills Program (CASP) checklist was used to assess the quality of the included studies.
The thesis revealed three aspects of opportunities of implementation AI in medical imaging. AI can enhance diagnostic accuracy and efficiency through its advanced capabilities. It also optimizes workflow by automating image acquisition, triage, and reporting process. For the next generation, AI even advances radiological education through adaptive learning and interactive simulation tools. Nevertheless, the thesis also identified two primary aspects of challenges when AI was applied to medical imaging. Many AI tools are trained on limited data, which can lead to biased results. There are also big ethical dilemmas regarding accountability and bias, and inadequate legal and regulatory frameworks. Furthermore, the absence of external validation and transparency in AI model development limits trust and clinical adoption.
These results can be concluded that AI has presented transformative potential for improving the quality, accessibility, and efficiency of medical imaging. However, its successful and equitable integration requires robust governance frameworks, transparent algorithmic design, and multidisciplinary collaboration to ensure ethical deployment. By addressing data and regulatory shortcomings, AI can become a cornerstone of a more patient-centred and efficient healthcare system.
This study aims to systematically explore the opportunity arising from the integration of AI in medical imaging and to identify the specific challenges healthcare organizations face in its deployment as well. The purpose of this study is to provide a synthesized understanding of how AI can be effectively and ethically utilized in medical imaging by integrating existing research data.
An umbrella review methodology was employed to collect and analyse evidence from 15 systematic review published between 2019 and 2024. Articles were collected from PubMed and ScienceDirect using defined inclusion and exclusion criteria. Data extraction and content analysis were conducted to identify recurring themes related to AI opportunities and challenges. The Critical Appraisal Skills Program (CASP) checklist was used to assess the quality of the included studies.
The thesis revealed three aspects of opportunities of implementation AI in medical imaging. AI can enhance diagnostic accuracy and efficiency through its advanced capabilities. It also optimizes workflow by automating image acquisition, triage, and reporting process. For the next generation, AI even advances radiological education through adaptive learning and interactive simulation tools. Nevertheless, the thesis also identified two primary aspects of challenges when AI was applied to medical imaging. Many AI tools are trained on limited data, which can lead to biased results. There are also big ethical dilemmas regarding accountability and bias, and inadequate legal and regulatory frameworks. Furthermore, the absence of external validation and transparency in AI model development limits trust and clinical adoption.
These results can be concluded that AI has presented transformative potential for improving the quality, accessibility, and efficiency of medical imaging. However, its successful and equitable integration requires robust governance frameworks, transparent algorithmic design, and multidisciplinary collaboration to ensure ethical deployment. By addressing data and regulatory shortcomings, AI can become a cornerstone of a more patient-centred and efficient healthcare system.