Reliability And Validity of 29 Questions Related to Perception of Radiologists Regarding Application of Artificial Intelligence in Radiology and Imaging Units of Tertiary Public Hospitals

Abstract

Background: Artificial intelligence (AI) is progressively being adopted in radiology, enhancing diagnostic precision, workflow efficiency, and image interpretation. For a developing country like Bangladesh, integrating AI into radiology is essential to advance radiology departments. Therefore, this study aimed to evaluate the validity and reliability of a 29-item questionnaire developed to evaluate radiologists’ perceptions regarding artificial intelligence applications in radiology and imaging units of tertiary public hospitals in Bangladesh Methodology: A cross-sectional methodological study was conducted among radiologists working in tertiary public hospitals. Validity was evaluated through face validity, by review from the research supervisor and subject experts to ensure clarity, relevance, and representativeness of the items. Content validity was established by confirming that the questions comprehensively covered the key domains of radiologists’ perception of AI and also by review from the research supervisor and construct validity was assessed by doing spearman rank correlation. Reliability of 29 questions was measured through internal consistency using Cronbach’s alpha. Results: The 29 questions which was designed to assess radiologist`s perception regarding application of AI was subjected to face validity through expert review and pilot testing to ensure clarity and relevance. Content validity was established by ensuring the 29 questions adequately covered key domains of radiologists’ perception of AI in radiology and imaging units. Construct validity of the questionnaire was supported by significant positive Spearman correlations between the domains of practice, opportunities, and challenges and the overall perception of radiologists. So, 29 questions were valid to assess radiologists` perception regarding application of AI in radiology and imaging units of tertiary public hospitals in Bangladesh. The 29 questions demonstrated good internal consistency. Subsection reliability was 0.832 for practice, 0.885 for opportunities, and 0.579 for challenges related to AI in radiology. Overall Cronbach’s alpha of 0.761 So, 29 questions related to perception of radiologists regarding application of AI were reliable. Conclusion: The developed 29-items in questionnaire was valid and reliable tool to measure radiologists’ perceptions regarding application of AI in radiology and imaging units of tertiary public hospitals. This instrument can facilitate further research and guide policymakers in addressing challenges and opportunities associated with AI adoption in radiology practice. JOPSOM 2025; 44(1): 24-30

Ähnliche Arbeiten