Patient at the centre and doctor as the ultimate decision maker: the decalogue for the correct use of AI in healthcare

First of all, the first 'commandment': Artificial Intelligence can indeed enhance the therapeutic relationship, giving back time and attention to the patient, but first of all it must be reiterated that the person is always at the centre. Just as it must always be remembered that AI is above all a support tool, rather than a substitute for clinical judgement, and therefore the 'ultimate decision-maker' in treatment must always and only remain the healthcare professional. But also that all results should always be 'explainable, verifiable and transparent'. These are the first three points of the valuable Decalogue drawn up by Anitec-Assinform, the Confindustria association of ICT companies, which together with the Multidisciplinary Working Table that brings together institutions, clinicians, companies, experts and stakeholders in the healthcare sector has also drawn up Guidelines for the adoption of Artificial Intelligence in the relationship between healthcare professional and patient. These documents were officially presented at an event on 11 December at the Chamber of Deputies, and which the companies are making available to the entire world of healthcare.

According to estimates, globally, the market for artificial intelligence applied to healthcare continues to grow rapidly: by 2025 it is expected to exceed USD 28 billion, up from around USD 1 billion in 2016. In Italy, an increase of more than 35% is expected as early as 2025 compared to 2024, making AI the technology with the highest growth rate among those currently used in healthcare. Today, the guidelines developed by Anitec-Assinform recall, the applications of AI in healthcare are already 'multiple and rapidly developing'. "Innovative technologies are improving operational efficiency, freeing up valuable time for healthcare professionals and allowing them to focus on the care relationship. They enable faster and more accurate diagnoses, as in the case of diagnostic imaging, where AI can support the physician in detecting early abnormalities that are difficult to detect with the naked eye, helping to improve the accuracy and effectiveness of the diagnostic process," the guidelines warn. They also point out that AI-enabled technologies also improve the patient care experience "in terms of efficiency, appropriateness and autonomy". However, in order to take full advantage of these opportunities, it is crucial that healthcare personnel are "adequately prepared for the introduction of these new technologies". This is also the starting point for the recommendations contained in the guidelines, a document that has been signed by some of the most representative associations and scientific societies in the healthcare world: FIMP, FNOPI, SID, SIIAM, SIMIG and SNAMI, as well as CONFCOOPERATIVE SANITA' and Cittadinanzattiva.

In particular, the document defines the guidelines for integrating AI in patient care and assistance, with a focus on the centrality of the person, the enhancement of the role of the healthcare professional, and the need for a conscious, safe and transparent adoption of digital technologies. The publication is part of an advanced stage of the transformation process started with Mission 6 Health of the NRP, which has decisively accelerated the digitisation of the NHS. In recent years, significant progress has been made throughout the country thanks to the advancement of the Plan's key projects, such as the Electronic Health Record 2.0, the Health Data Ecosystem, the National Telemedicine Platform, and the Artificial Intelligence Platform to support primary care. The work highlights how AI can become an ally of the healthcare professional, able to reduce the administrative burden, improve the organisation of clinical information, including through the assisted compilation of the patient summary, and support decision-making without ever replacing clinical judgement. "Experience teaches us that technology realises its full potential when it is understood, accepted and used consciously by its users. This is why the guidelines we are presenting today stem from an ongoing dialogue between companies, institutions and third parties representing doctors, healthcare professionals and patients. AI can support the evolution of the National Health Service, but only if it is embedded in processes that are clear, reliable and truly oriented towards the needs of people and structures,' Massimo Dal Checco, President of Anitec-Assinform.

In addition to the centrality of the patient and AI as a support for healthcare and not as a substitute tamong the priorities emphasised by the document are also the interoperability of systems, data quality, the reduction of digital fragmentation, ongoing training of professionals, technology assessment and validation paths, privacy protection, and participatory governance involving all healthcare system actors. The Guidelines also emphasise how Artificial Intelligence can help reduce territorial inequalities, expanding access to digital services even in inland and less infrastructured areas. In conjunction with the document, a concise and easy Decalogue for the adoption of AI in the relationship between health professional and patient has been produced, which summarises in ten principles the cornerstones of the proposal: centrality of the therapeutic relationship, territorial equity, responsibility and safety, transparency, the role of the professional as the ultimate decision-maker, interoperability of data, multi-stakeholder co-design, protection of privacy and consent, appropriateness and impact assessment. These guidelines and the Decalogue therefore represent for Anitec-Assinform an important step towards a more modern, interoperable healthcare system centred on the relationship between professional and patient, offering concrete guidance to institutions, healthcare facilities, professionals and businesses in the crucial digital transition phase that the country is currently going through.