Automating Clavien–Dindo classification with large language models in percutaneous nephrolithotomy: an exploratory study
Gustavo Perrone, Alexandre Danilovic, Carlos Batagello, Fabio Cesar Miranda Torricelli, Fabio Carvalho Vicentini, Giovanni Scala Marchini, Rodrigo Perrella, William C. Nahas, Eduardo Mazzucchi · 2026
Read Automating Clavien–Dindo classification with large language models in percutaneous nephrolithotomy: an exploratory study by Gustavo Perrone, Alexandre Danilovic, Carlos Batagello, Fabio Cesar Miranda Torricelli, Fabio Carvalho Vicentini, Giovanni Scala Marchini, Rodrigo Perrella, William C. Nahas, Eduardo Mazzucchi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13619659.txt.
Description
Gustavo Perrone and eight colleagues tested whether large language models can grade surgical complications as reliably as doctors. The setting was percutaneous nephrolithotomy, a kidney stone operation done through the skin. Six chatbots, including ChatGPT, Gemini, and DeepSeek, were asked to define the Clavien-Dindo grading scale, grade complication descriptions, and grade case summaries, with and without example prompts.
Every model defined the scale correctly and agreed with the reference grades almost perfectly. Six attending urologists and fifteen residents grading the same cases agreed slightly less, and the fastest doctor still needed seventeen minutes while models finished in a few minutes at most. Agreement among the chatbots also beat agreement among physicians. The authors call the result promising for an artificial setting, and they ask for prospective tests on real records plus debate about legal and ethical limits.
Catalog record
| CREATOR | Gustavo Perrone, Alexandre Danilovic, Carlos Batagello, Fabio Cesar Miranda Torricelli, Fabio Carvalho Vicentini, Giovanni Scala Marchini, Rodrigo Perrella, William C. Nahas, Eduardo Mazzucchi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |