For psychiatrist Fernando Corinto, AI is a valuable tool for research and clinical practice, but diagnosis remains an act of professional responsibility.
Artificial intelligence can analyse vast amounts of data, identify correlations and support clinical work. But when it comes to making a diagnosis, interpreting a complex case or building a therapeutic relationship, the responsibility remains with the professional. This is the conviction of Fernando Corinto, a psychiatrist specialising in the study of autism in adolescents and adults, who took part in the training programme The Future of Care. In this interview, conducted by Onelia Onerati, Corinto explains how AI is already changing the way he studies, writes and conducts research, without altering what he considers essential in clinical practice: judgement, the ability to ask the right questions and a focus on the individual, in their irreducible uniqueness.
A self-employed practitioner, he completed his studies in Psychiatry at the University of Tor Vergata in Rome. “Psychiatry historically emerged as a discipline called upon to examine and explain madness – that is, those forms of mental suffering which, due to their intensity and peculiarity, found no place in other fields of medicine. Over time, however, its scope has gradually expanded to encompass the entire spectrum of mental disorders and conditions of psychological distress, addressing not only the most severe and marginal manifestations, but also the emotional, relational and behavioural difficulties that affect the lives of many people. “My specialist area of interest focuses primarily on autism in adolescents and adults, but I treat patients with all other mental disorders,” he explains.
Fernando is in his early thirties and began his career as a psychiatrist at a time when the use of technology was already well established in medicine. “Electronic health records and automated reporting systems were already established practice when I began practising. This is one of the reasons why I have always been convinced that innovation must be fully integrated; I am thinking in particular of the study and research phase in neuroscience, where there are as yet no appreciable clinical results. In particular, technology in psychiatry can be fundamental in supporting functional magnetic resonance imaging (fMRI) studies that investigate the correlates of thought disorders. It is precisely on this front that a link is established between the ‘organic’ component (the brain) and the ‘inorganic’ component (thought), capable of explaining how drugs work at a specific level, where they act and why they work.”
What role might artificial intelligence play in this delicate research? “AI enables us to carry out data analysis that is crucial for identifying correlations between factors. In Scandinavia, for example, millions of data points—meticulously collected from both patients and the general population—are already available. Analysing such a vast amount of information can be challenging, and AI enables us to do so, either confirming or refuting our intuitions.”
What has changed in your use of AI since you began working on the material for *The Future of Care*? “I have only recently completed the course and have started using the tools more extensively, for example to consult various guidelines from verified sources.
AI helps me when studying medicines: how they interact with other medicines, especially in the case of newer drugs; I also receive a great deal of support with writing. It is much quicker and easier to draft reports tailored to the recipient (from the patient to a colleague) and to present them in a more visually appealing and functional format.”
Everything proved to be extremely user-friendly: “The content of ‘The Future of Care’ is presented clearly, with key aspects well highlighted and analysed comprehensively from multiple perspectives. I appreciated the modular structure and the ability to rewatch the videos, as well as the examples and use cases”.
Taking part in such a structured course prompts a great deal of reflection: “I’ve given more thought to the issue of privacy; one must always take the utmost care when using anonymised information because one does not have the patient’s authorisation. Furthermore, it is important to always remember that, when evaluating AI outputs, one must not abdicate responsibility for the report. Whatever the result of an AI analysis, it is the doctor who signs off on the diagnosis! Furthermore, we must always pay close attention to the analyses provided: how much information is simply overlooked for no real reason!”
Finally, there were various reflections on the doctor-patient relationship. “On the one hand, thanks to these tools, the time spent on administrative tasks is greatly reduced compared to clinical work, and the relationship with the people in front of us is given more space. We do, in fact, consider filling in a medical record to be very demanding, as it has legal standing. On the other hand, it is important to rediscover the role we doctors had before the advent of the internet: to re-establish the therapeutic relationship, to be the one who interprets the test results, and to clarify any doubts and questions. Nowadays, all patients come to us convinced that they already know their diagnosis. So we must not be afraid to question the algorithm, but rather guide the patient through the AI’s responses, interpreting and mediating them. In my field, this process is even more delicate. Thinking of my patients, I am aware that I must work with great care so as not to undermine the therapeutic relationship. Everyone wants to put a label on their problem, and AI provides immediate answers without truly weighing up the clinical case before it – a case that always has its own intrinsic singularity and uniqueness.”
