Table 2.

The assessment of the process factors¹

1. Regarding the maturity of the model, is it still in the retrospective development phase or does it already have approval for implementation and prospective evaluation?	“Yes, it has already been approved with the Ethical Committee and AOUPR (Azienda Ospedaliero-Universitaria di Parma) but is not yet deployed for clinical use”
2. Did you use a multidisciplinary approach involving experts from different fields?	“Yes, we work with a multidisciplinary team composed not only of Anaesthesia and Intensive Care specialists, but also of software, clinical and management engineers.”
3. Did you use a so-called ‘devil’s advocate’ attitude to counterbalance the hype and excessive optimism, e.g. by including people in the team who were sceptical about AI applications?	“Yes, we have been working with both Unipr and AOUPR engineers. In the project group there are both those who obviously believe in AI and those who were potentially sceptical and decided to join once they got to know the project better.”
4. Have you already prepared guidelines for the actual implementation of the model?	“Not yet, but we will try to follow an implementation”
5. Assessment should be done on a regular basis during the AI deployment phase, so when should the assessment be revisited?	“At date, we regularly check the IoT equipment. If it comes to implementation, there will certainly be regular checks of the whole system.”

¹ the MAS-AI model recommends a broader update; not just at check of the AI-models performance or the technical aspects, but to revisit the entire evaluation