Artificial Intelligence Integration in Radiology in Saudi Arabia: A Cross-Sectional Survey of Workforce Readiness and Implementation Challenges

Background: Artificial intelligence (AI) is increasingly being integrated into radiology and may improve diagnostic efficiency and workflows. Its successful adoption depends on workforce readiness, organizational capacity, and effective governance. This study evaluated radiology professionals’ perceptions of AI integration in Saudi Arabia; sustainability was considered only as a conceptual implementation context and was not directly measured. Methods: An online cross-sectional survey using convenience sampling was conducted among healthcare professionals involved in radiology services in Saudi Arabia, using an expert-reviewed and pilot-tested 30-item questionnaire covering knowledge, attitudes, implementation readiness, and perceived barriers. Responses from 295 healthcare professionals were analyzed using descriptive statistics, Cronbach’s alpha, Spearman correlation, and Kruskal–Wallis testing. Results: Participants demonstrated positive attitudes toward AI integration (3.57 ± 0.69) and moderate knowledge (3.35 ± 0.70), whereas implementation readiness was comparatively lower (3.04 ± 0.79). Perceived barriers showed the highest domain score (3.67 ± 0.64). Major barriers included implementation costs (3.98 ± 0.71), limited digital infrastructure (3.90 ± 0.75), insufficient staff training (3.84 ± 0.77), and lack of technical expertise (3.82 ± 0.78). Positive correlations between knowledge, attitudes, and implementation readiness were observed. Conclusions: Among participating respondents, support for AI integration was generally favorable, but lower perceived institutional readiness and organizational, infrastructural, and governance barriers may constrain implementation. Because the convenience sample lacked a sampling frame and a calculable response rate, the results should not be interpreted as national estimates. The survey also did not measure clinical, economic, resource-use, or environmental outcomes and therefore does not establish sustainability benefits. Future probability-based and implementation studies should evaluate representativeness and these outcomes directly.