We evaluated an AI chatbot's ability to suggest diagnostic and therapeutic pathways for renal cell carcinoma (RCC) in a multidisciplinary tumor board (MDT). A retrospective analysis of 103 cases (2023-2024) found 62.1% agreement with MDT decisions (kappa = 0.44, p< 0.001). Concordance was highest in when follow-up imaging was suggested (p = 0.001), with disease status influencing agreement (p = 0.004). These results suggest AI could assist in RCC case assessments, warranting further research.
Role of large language models in the multidisciplinary decision-making process for patients with renal cell carcinoma: a pilot experience
Bertolo, Riccardo
;De Bon, Lorenzo;Caudana, Filippo;Pettenuzzo, Greta;Malandra, Sarah;Casolani, Chiara;Fantinel, Emanuela;Borsato, Alessandro;Negrelli, Riccardo;Brunelli, Matteo;Cerruto, Maria Angela;Antonelli, Alessandro
2025-01-01
Abstract
We evaluated an AI chatbot's ability to suggest diagnostic and therapeutic pathways for renal cell carcinoma (RCC) in a multidisciplinary tumor board (MDT). A retrospective analysis of 103 cases (2023-2024) found 62.1% agreement with MDT decisions (kappa = 0.44, p< 0.001). Concordance was highest in when follow-up imaging was suggested (p = 0.001), with disease status influencing agreement (p = 0.004). These results suggest AI could assist in RCC case assessments, warranting further research.
File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
