Artificial Intelligence in Radiology: A Cross-sectional Study to Assess the Awareness, Acceptance and Anticipated Challenges among Postgraduate Students

Abstract

Introduction: The article aims to assess the existing awareness and attitude of postgraduate residents toward the use of artificial intelligence (AI) in radiology practice. The acceptance of AI is of paramount importance before it becomes part of future practice. Methods: The study is a cross-sectional survey among Radiology post graduate students in India. The students were MD or DNB residents from parts of India. The study population comprised of almost equal distribution from government and private institutions. A total of 108 radiology residents participated in the study. Questionnaires to postgraduate residents were shared and their responses were recorded. Statistical analysis was used to critically study their responses and the conclusions were made. Results: The questionnaire were based to study the demographics, familiarity and perceptions on artificial intelligence in radiology, concerns about artificial intelligence’s impact, clinical and ethical implications of artificial intelligence, responsibility and ethical use of artificial intelligence. The awareness regarding AI is good among the postgraduate residents radiology residents. A majority (75%) felt radiologists should not be held responsible for AI induced errors, indicating a view that AI tools should complement, not replace, human oversight. Conclusion: There is widespread optimism about AI’s potential to enhance diagnostic accuracy and clinical efficiency. There are a few notable concerns around its potential to displace jobs. Nearly 44% of participants agreed with the statement, AI will not replace radiologists, AI will replace radiologists without AI. The majority of participants support the adoption of AI‐enabled equipment, underscoring a readiness to integrate technological advancements into their practice. The mixed feelings about AI’s potential influence highlight an urgent need for structured training programs and upskilling initiatives, ensuring that radiologists can responsibly and effectively utilize AI tools. The responses reflect a prevailing sentiment that AI is an essential tool, not a standalone solution, reinforcing the importance of human AI collaboration.

Ähnliche Arbeiten