KneeXNeT: An Ensemble-Based Approach for Knee Radiographic Evaluation
Nicharee Srikijkasemwat, Soumya Snigdha Kundu, Fuping Wu, Bartłomiej W. Papież · 2024
Read KneeXNeT: An Ensemble-Based Approach for Knee Radiographic Evaluation by Nicharee Srikijkasemwat, Soumya Snigdha Kundu, Fuping Wu, Bartłomiej W. Papież (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/arxiv-2412-07526.txt.
Description
Nicharee Srikijkasemwat and co-authors describe a 2024 system that grades knee osteoarthritis from X-ray images. The disease is the most common joint disorder and a leading cause of disability. Grading its severity by eye, on the Kellgren-Lawrence scale, takes expert time that clinics often lack.
The team first tested ten modern deep learning models, and the best single model scored about sixty-nine in one hundred. Weighted sampling to fix class imbalance nudged that score slightly higher. Heat maps from Smooth-GradCAM++ showed which image regions drove each call, adding trust. Joining models into an ensemble, named KneeXNeT, lifted accuracy to 0.72, the best mark in the study. The authors offer it as an aid that could ease expert workload.
Catalog record
| CREATOR | Nicharee Srikijkasemwat, Soumya Snigdha Kundu, Fuping Wu, Bartłomiej W. Papież |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 15 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-07526