Intersection of Performance, Interpretability, and Fairness in Neural Prototype Tree for Chest X-Ray Pathology Detection: Algorithm Development and Validation Study

Hongbo Chen, Myrtede Alfred, Andrew D Brown, Angela Atinga, Eldan Cohen · 2024

Read Intersection of Performance, Interpretability, and Fairness in Neural Prototype Tree for Chest X-Ray Pathology Detection: Algorithm Development and Validation Study by Hongbo Chen, Myrtede Alfred, Andrew D Brown, Angela Atinga, Eldan Cohen (2024) 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-11659703.txt.

Description

Hongbo Chen, Myrtede Alfred, Andrew Brown, Angela Atinga, and Eldan Cohen build a chest X-ray reader that can show its reasoning. Deep learning models read scans well but rarely explain a call, which slows their use in clinics. The team's model blends a neural net with a decision tree built from typical image patches. They trained it on three large public scan sets and checked accuracy, clarity, and fairness.

Clearer models paid a price in accuracy, and the balance shifted by disease. The paper shows both single-scan and whole-model explanations. It weighs how much clarity a clinic can afford before fairness or accuracy suffers.

Catalog record

CREATORHongbo Chen, Myrtede Alfred, Andrew D Brown, Angela Atinga, Eldan Cohen
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11659703

Browse the full library · About the Tower of Babel