Radiology Residents' and Radiologists' Perception and Attitude Towards Medical Artificial Intelligence in Radiology– An Initial National Multicenter Survey

Abstract Introduction : Recent advances in artificial intelligence and machine learning (AI/ML) are transforming radiology practices. While AI/ML innovations present opportunities to augment radiologists' capabilities, some have expressed concerns about AI/ML potentially replacing radiologists in the future. These uncertainties have led to varied perspectives among radiology professionals regarding the role of AI/ML in the field. This study aimed to assess respondents' knowledge, research involvement, utilization of AI/ML applications, and attitudes towards the impact of AI/ML on radiology practice and training. Methods Between June and July of 2022, we conducted a web-based survey of radiologists and radiology residents from 5 major institutions in Ethiopia with radiology residency programs. The survey was distributed through the Ethiopian Radiological Society, and social media. Group comparison was tested by chi-square test for categorical responses and Mann-Whitney test for ordinal rating scale responses. Results Of the 276 respondents, 94.5% were novices when it came to AI/ML, and radiologists were more likely than residents to have read a journal paper on AI in radiology in the previous 6 months (33.3% vs. 18.9%). Only 1.8% of respondents had active or previous involvement in AI research, though 92% were eager to join such research efforts. Most of respondents intended to expand their AI/ML knowledge (84.6%) and believed AI/ML would substantially influence radiology practice (72.3%). While few felt AI/ML could replace radiologists (16.8%), most supported integrating AI/ML training into radiology residency curricula (82.9%). Conclusion This study suggests that radiology residents and radiologists in Ethiopia are generally positive and open-minded towards AI/ML in radiology, despite their limited knowledge and experience with the technology. The majority of respondents believe that AI and data science skills should be introduced during residency training. Recommendations : Medical AI training should be incorporated into radiology residency programs to prepare future radiologists for the changing landscape of radiology practice.