3D ReX: Causal Explanations in 3D Neuroimaging Classification

Melane Navaratnarajah, Sophie A. Martin, David A. Kelly, Nathan Blake, Hana Chockler · 2025

Read 3D ReX: Causal Explanations in 3D Neuroimaging Classification by Melane Navaratnarajah, Sophie A. Martin, David A. Kelly, Nathan Blake, Hana Chockler (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/arxiv-2502-12181.txt.

Description

Melane Navaratnarajah and four co-authors explain 3D brain scan classifiers with causal maps. Their 2025 paper introduces 3D ReX, a tool for opening black-box medical AI. Doctors hesitate to trust scan verdicts they cannot inspect. Earlier explainers worked on flat images, not full volumes.

The tool builds on the theory of actual causality. It assigns responsibility scores to regions of the scan. The brightest regions are those the verdict truly hinged on. A stroke detection model served as the test case. The maps showed where stroke signs mattered most in the volume. Causal maps could help clinicians judge when to trust the machine.

Catalog record

CREATORMelane Navaratnarajah, Sophie A. Martin, David A. Kelly, Nathan Blake, Hana Chockler
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-12181

Browse the full library · About the Tower of Babel