Keeping AI in medicine and radiology within the framework of scientific method: measuring to close the epistemic gap

Filippo Pesapane, Francesco Sardanelli · 2025

Read Keeping AI in medicine and radiology within the framework of scientific method: measuring to close the epistemic gap by Filippo Pesapane, Francesco Sardanelli (2025) 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/pmc-12722631.txt.

Description

Filippo Pesapane and Francesco Sardanelli argued in 2025 that artificial intelligence in medicine and radiology must stay inside the framework of the scientific method. Modern AI systems give answers end to end and explain them afterward, which can skip the middle steps of careful testing. Explanations may also miss the real basis of a model's choice.

Imaging studies show warning signs such as unstable saliency maps, shortcut learning, and failure when data shift. Training on the local patient group can help fit but may lock in local bias. The authors propose hospital-run programs with tracked data, preregistered tests, checks for causal alignment, published failures, and both local and outside validation.

Catalog record

CREATORFilippo Pesapane, Francesco Sardanelli
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE17 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12722631

Browse the full library · About the Tower of Babel