The knowledge and perception of patients in Malta towards artificial intelligence in medical imaging

Abstract

Purpose: To investigate the current knowledge of AI amongst patients in Malta and determine how patients in Malta perceive the application of AI in medical imaging. Methodology: A cross-sectional, prospective, non-experimental study using a quantitative approach was adopted. The self-designed paper-based questionnaire, partly adapted with permission from a previous survey and questionnaire, was distributed in English or Maltese language amongst eligible outpatients attending medical imaging examinations in Maltese or Gozitan hospitals in March 2023. Results: 280 completed questionnaires were analysed, yielding a 5.83% confidence interval. 42.1% of patients indicated basic knowledge of AI, while 36.4% reported minimal to no knowledge. The responses revealed that patients were mostly favourable to humans and AI complementing each other to improve healthcare delivery. Nevertheless, most patients indicated a strong preference for doctors being involved and retaining the final-decision making, even if AI is used. Furthermore, most considered radiologists’ replacement unlikely. For some statements, a statistically significant association was noted between patients’ perception of AI-based technology and their gender, age, educational background, and technological knowledge. Specifically, females, older patients, and those with lower educational levels and limited technological skills, demonstrated less trust in AI and its ability to recommend reliable treatment options. Most patients (92.1%) expressed the desire to be informed if AI was used in their care, with a considerable percentage of respondents (76.8%) indicating their belief that AI integration with medical imaging could improve healthcare delivery. Conclusions: Patients are important stakeholders in the development of AI technology, and therefore, their involvement is crucial for prioritising patient-centred care, addressing ethical concerns, fostering trust, and ensuring AI systems align with their values and needs. Implications for practice: Educational programs are necessary to improve AI literacy amongst patients. Additionally, clear policies on AI use in radiography must be implemented to safeguard patients’ rights and well-being.