Artificial intelligence (AI) is rapidly transforming radiography and medical imaging in several aspects of practice to support more sustainable healthcare delivery: workflow optimization, image acquisition, image reconstruction, image interpretation, workflow optimization, clinical decision support, and radiation dose management. However, questions regarding its environmental impact continue to persist.
AI models contribute significantly to carbon emissions, electronic waste, and resource consumption due to the energy-intensive processes involved in training, deploying, and maintaining these models. Thus, understanding how AI can improve contemporary radiological practice while minimizing its ecological footprint is rising to prominence and becoming an important research priority for imaging professionals, academics, clinicians, policymakers, healthcare managers, and industry stakeholders.
This issue explores the dual forces in the concepts, Green AI and AI for Green.
Green AI emphasizes the development of environmentally sustainable AI technologies that minimizes energy consumption, carbon emissions and other environmental impacts while maintaining acceptable clinical performance.
AI for Green on the contrary focuses on using AI to promote sustainability and reduce environmental burdens within healthcare systems such as streamlining clinical workflows, improving resource utilization, reducing repeat imaging, radiation dose optimization, and supporting environmentally responsible radiography services. The primary goal is to advance interdisciplinary scholarship and evidence-based practice on the environmental sustainability of AI applications in radiography.
We invite contributions that evaluate the carbon footprint, energy efficiency, clinical effectiveness, economic implications, and environmental value of AI across radiography integrated interdisciplinary pathway; as well as explore the opportunities, challenges, ethical considerations and practical solutions associated with balancing AI-driven innovation and environmental stewardship within radiological sciences. We welcome researchers from various disciplines to provide interdisciplinary perspectives. Your contributions will play a crucial role in advancing knowledge in this field.
Potential topics include, but are not limited to:
- Carbon footprint assessment and lifecycle analysis of AI applications in radiography
- Energy-efficient AI algorithms, machine learning and deep learning models for diagnostic imaging, lightweight models, model compression, and sustainable computing
- Green AI frameworks, sustainable computing approaches in healthcare and sustainability metrics, reporting frameworks, and methods for assessing the environmental performance of radiology AI
- Comparative evaluations of Green AI and AI for Green strategies in medical imaging
- AI-driven radiation dose optimisation and reduction of repeat examinations
- AI-enabled workflow efficiency, equipment utilisation, and resource optimisation
- Environmental impacts of cloud-based AI, data centres, medical imaging archives, data-intensive healthcare systems and edge computing
- Ethical, regulatory, and governance considerations surrounding Green AI in healthcare
- Trade-offs between diagnostic accuracy, computational efficiency, patient safety, and environmental sustainability
- Economic evaluations, implementation barriers, and policy frameworks for sustainable AI adoption
- Education, workforce development, professional awareness, and organisational strategies for environmentally sustainable and responsible AI in radiography
- Future directions for environmentally responsible AI innovation in medical imaging
