Since its inception, the internet has provided digital solutions that have impacted every social sphere, from daily life to the working world. This digital revolution has also had repercussions in healthcare, with the emergence of e-health. In recent years, the development of artificial intelligence and large language models (LLMs) – algorithms based on deep neural networks capable of learning from vast amounts of textual and non-textual data – has breathed new life into unexplored peaks of healthcare digitalization. This has also led to an evolution in the work of healthcare professionals, including nurses, who find themselves at the forefront of receiving and disseminating these new digital solutions. This thesis aims to conduct a scoping review, divided into four chapters, analyzing the potentialities and practical applications of large language models in healthcare, specifically focusing on ChatGPT – a chatbot created by the company OpenAI capable of fully leveraging the potential of LLMs – and its application in clinical nursing. Specifically, the first chapter will provide an overview of the state of the art in the digital healthcare revolution. We will then delve into the topic of artificial intelligence, outlining the fundamental steps that have led to its application in healthcare. Finally, we will illustrate the application of ChatGPT in the clinical healthcare domain, focusing on its potential for various stakeholders in healthcare and the complexities encountered in its use. The subsequent chapters will be dedicated to specifying the materials and methods used for this study, ultimately presenting results and conclusions.
Sin dalla sua nascita, internet ha fornito soluzioni digitali che hanno impattato ogni campo sociale, dalla vita quotidiana al mondo del lavoro. Questa rivoluzione digitale ha avuto ripercussioni anche in sanità, con la nascita dell’e-health. Negli ultimi anni, lo sviluppo di intelligenze artificiali e large language models (LLM) – algoritmi basati su reti neurali profonde in grado di apprendere da enormi quantità di dati testuali e non – ha dato nuova linfa per raggiungere vette ancora inesplorate nella digitalizzazione sanitaria. Tutto ciò ha portato anche un’evoluzione nel lavoro dei professionisti sanitari, tra cui l’infermiere, che si ritrova ad essere uno dei punti cardine per la ricezione e la diffusione di queste nuove soluzioni digitali. Con questo elaborato di tesi, suddiviso in quattro capitoli, si intende effettuare una scoping review, analizzando le potenzialità e le applicazioni pratiche dei large language models in sanità, con riferimento specifico a ChatGPT – chatbot creato dalla società OpenAI in grado di sfruttare a pieno le potenzialità dei LLM – e alla sua applicazione nell’ambito infermieristico clinico. Per la precisione, nel primo capitolo si fornirà una panoramica sullo stato dell’arte della rivoluzione digitale sanitaria. Si procederà, successivamente, ad inoltrarci nel tema delle intelligenze artificiali, delineandone i passaggi fondamentali che hanno portato, alla loro applicazione in sanità. Procederemo, infine, con l’illustrare l’applicazione di ChatGPT in ambito clinico sanitario, con un focus sulle potenzialità per i vari attori coinvolti nell’healthcare e alle complessità di utilizzo riscontrate. I successivi capitoli saranno dedicati a specificare i materiali e metodi utilizzati per tale studio, declinando infine risultati e conclusioni.
Large language models e ChatGPT: una scoping review per evidenziare potenzialità ed applicabilità nell'ambito infermieristico
VARAVALLO, SALVATORE
2022/2023
Abstract
Since its inception, the internet has provided digital solutions that have impacted every social sphere, from daily life to the working world. This digital revolution has also had repercussions in healthcare, with the emergence of e-health. In recent years, the development of artificial intelligence and large language models (LLMs) – algorithms based on deep neural networks capable of learning from vast amounts of textual and non-textual data – has breathed new life into unexplored peaks of healthcare digitalization. This has also led to an evolution in the work of healthcare professionals, including nurses, who find themselves at the forefront of receiving and disseminating these new digital solutions. This thesis aims to conduct a scoping review, divided into four chapters, analyzing the potentialities and practical applications of large language models in healthcare, specifically focusing on ChatGPT – a chatbot created by the company OpenAI capable of fully leveraging the potential of LLMs – and its application in clinical nursing. Specifically, the first chapter will provide an overview of the state of the art in the digital healthcare revolution. We will then delve into the topic of artificial intelligence, outlining the fundamental steps that have led to its application in healthcare. Finally, we will illustrate the application of ChatGPT in the clinical healthcare domain, focusing on its potential for various stakeholders in healthcare and the complexities encountered in its use. The subsequent chapters will be dedicated to specifying the materials and methods used for this study, ultimately presenting results and conclusions.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/17340