Table 3.

Structured analysis of the BLOC-OP project according to the MAS-AI model criteria

STEP 1 – early MAS-AI
	Domain Description	BLOC-OP analysis
1	The health problem and current use of technology	The management of operating block is very complex and is currently imprecise and based on a very limited number of variables. Initial results in the literature regarding the use of AI for this purpose were promising. Therefore, it was decided to test the use of AI models to improve surgical scheduling and simultaneously exploit an indoor tracking system to optimize the data quality and reliability.
2	Technology	The first phase consisted of the collection of a solid dataset, also exploiting indoor tracking system for the direct extrapolation of surgical and recovery room times. The second phase was the application of AI techniques to be able to automatically process the data and provide models for predicting operating room and recovery room occupancy and then scheduling surgical procedures.
3	Ethical aspects	The study was conducted in accordance with Good Clinical Practice (GCP) standards and the Declaration of Helsinki.
4	Legal aspects	The authors consulted with a team of experienced AI lawyers to ensure compliance with the regulatory framework and to adapt the informed consent forms.
STEP 2 – full MAS-AI
5	Safety	Security measures were designed together with software engineers. BLOC-OP technology adopts several security measures such as WP2 encryption, Media Access Control (MAC) address constraints, the use of MongoDB databases and RAID media to guarantee data protection, limit access only to authorised devices and ensure the reliability and availability of information.
6	Clinical aspects	The clinical aspects are mainly related to the fact that a better optimisation of resources leads to an increase in the quality and safety of the services provided.
7	Economic aspects	The economic impact of the new system is among the secondary objectives of the project and to be investigated later on. More accurate scheduling is expected to lead to an optimisation of available resources.
8	Organisational aspects	The ultimate goal is to create a technology that takes into account all the information relevant to this decision-making process, and that is autonomous in making the most effective plan. In addition, the model will be designed to be easily implemented in daily clinical practice.
9	Patient aspects	The technology does not place patients at any additional risk. Moreover, in the process of collecting informed consent, patients are given information about the use of AI techniques.