MSCA-TNet based deep learning method for ECG arrhythmia classification
Songjian Huang, Senpeng Chen, Zihao Guo, Kaiming Ji, Miao Zhang · 2026
Read MSCA-TNet based deep learning method for ECG arrhythmia classification by Songjian Huang, Senpeng Chen, Zihao Guo, Kaiming Ji, Miao Zhang (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/PMC13554250.txt.
Description
Songjian Huang leads a five-author team building a model that reads heart tracings and sorts abnormal rhythms in a 2026 study. Wearable recorders gather these tracings in messy settings, so noise and motion corrupt the signal. Rare rhythm types also appear far less often than common ones, skewing training. The team cleaned signals with a wavelet filter and rebalanced the training data by blending rare cases.
Their network pairs an attention step, which stresses the most telling channels, with a transformer that reads long-range patterns. Local detail and whole-trace context thus meet in one judgement. On two public benchmarks the model scored near ninety-nine percent overall accuracy. Gains were strongest for the rare classes that older models often miss.
Catalog record
| CREATOR | Songjian Huang, Senpeng Chen, Zihao Guo, Kaiming Ji, Miao Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38405 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |