Which explainable AI methods in medical imaging are clinically impactful? A systematic literature review addressing the clinician's perspective

Shahab Ud Din, Ruby Kemna, Johannes C. F. Ket, Muhammad Iqbal, Omar Bohoudi, Mark Hoogendoorn, Elena Beretta, Aneta Lisowska · 2026

Read Which explainable AI methods in medical imaging are clinically impactful? A systematic literature review addressing the clinician's perspective by Shahab Ud Din, Ruby Kemna, Johannes C. F. Ket, Muhammad Iqbal, Omar Bohoudi, Mark Hoogendoorn, Elena Beretta, Aneta Lisowska (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/pmc-13260647.txt.

Description

Shahab Ud Din and colleagues reviewed in 2026 how clinicians, not just engineers, judge AI tools that explain their calls in medical imaging. Many explanation tools exist, but most tests measure technical quality rather than use at the bedside. Their search across five databases began with more than nine thousand records.

After screening, only 51 studies had truly asked health workers to rate the explanations. Example-based and concept-based tools tended to improve both how clinicians felt and how they performed. Methods that highlight image regions mostly improved feelings, with unclear effects on choices and a risk of over-trust. The authors judge the field promising but fragile in its methods. They call for shared, clinician-centered standards before these tools enter routine care.

Catalog record

CREATORShahab Ud Din, Ruby Kemna, Johannes C. F. Ket, Muhammad Iqbal, Omar Bohoudi, Mark Hoogendoorn, Elena Beretta, Aneta Lisowska
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE80 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13260647

Browse the full library · About the Tower of Babel