The use of the social robot NAO in medical settings: how to facilitate interactions between healthcare professionals and patients with autism spectrum disorder.

Objectives: This study investigates how to facilitate the use of the social robot NAO in medical settings to support interactions with children diagnosed with Autism Spectrum Disorder (ASD). The objective was to develop intuitive control methods that enable healthcare professionals to easily integrate the robot into clinical practice. Methods: Two control modes were designed: Puppet mode, where clinicians manually operate the robot via a graphical console, and Assistant mode, where a Large Language Model translates clinicians' spoken requests into robot actions and dialogue. Twenty-three doctors evaluated both modes through video demonstrations and completed questionnaires assessing usability, usefulness, and ethical acceptability. Results: Both modes were considered effective and user-friendly. Assistant mode was perceived as more intuitive and adaptable, facilitating seamless interaction, whereas Puppet mode was judged slightly more reassuring for patients and somewhat more appropriate in terms of robot actions. Conclusion: Overall, both approaches were positively received, with Assistant mode emerging as the preferred option for integration into clinical workflows due to its perceived simplicity and flexibility. These findings highlight clinicians' positive perceptions of two novel control modes and emphasize NAO's potential to enhance patient engagement and reduce stress. Further empirical validation with children in real clinical trials is warranted to confirm these benefits and optimize robot-assisted interventions in ASD care.