DCBM-Tri: a dual-channel bilinear mapping triplet model for early recognition of acute kidney injury in imbalanced cohorts

Kai Wang, Ling Lin, Xinye Jin, Han Chen, Xudong Lu, Huilong Duan, Shan Nan · 2026

Read DCBM-Tri: a dual-channel bilinear mapping triplet model for early recognition of acute kidney injury in imbalanced cohorts by Kai Wang, Ling Lin, Xinye Jin, Han Chen, Xudong Lu, Huilong Duan, Shan Nan (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/PMC13542066.txt.

Description

Kai Wang and six colleagues present, in a 2026 modelling study, a deep learning system for spotting acute kidney injury in intensive care patients twelve hours before it declares itself. Early warning is hard partly because injury cases are heavily outnumbered by stable patients, which dulls ordinary models.

Their dual-channel design learns by contrasting similar and dissimilar patients, sharpening the features that separate future cases. Tested on an imbalanced cohort, it beat conventional baselines including recurrent networks and resampling tricks on most measures, and offered net clinical benefit across a wider risk range. Explanation analysis elevated five routine measures, led by C-reactive protein and ionised calcium. Honest limits are noted where one comparison showed no clear gain.

Catalog record

CREATORKai Wang, Ling Lin, Xinye Jin, Han Chen, Xudong Lu, Huilong Duan, Shan Nan
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE60276
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13542066

Browse the full library · About the Tower of Babel