Comparing Deep Neural Network for Multi-Label ECG Diagnosis From Scanned ECG
Cuong V. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected], Hieu X. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected] Dung D. Pham MinhCollege of Engineering and Computer ScienceVinUniversity, Ha · 2025
Read Comparing Deep Neural Network for Multi-Label ECG Diagnosis From Scanned ECG by Cuong V. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected], Hieu X. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected] Dung D. Pham MinhCollege of Engineering and Computer ScienceVinUniversity, Ha (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-2502-14909.txt.
Description
Nguyen and co-authors compare deep learning models in a 2025 study for reading heart tracings from scanned paper. Clinics in many places still store ECGs as paper or scanned images, not as clean digital signals. Automated reading of such scans is harder than reading the raw signal, and most past work only split scans into normal and abnormal.
The team tests several well-known network designs, from AlexNet and VGG to ResNet and a vision transformer, on scanned ECG data. Their task assigns several labels at once, since one tracing can show more than one problem. They compare accuracy, strength against scan flaws, and how well each model carries to new cases. The study also checks whether scans keep enough detail for safe automated reading.
Catalog record
| CREATOR | Cuong V. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected], Hieu X. NguyenCollege of Engineering and Computer ScienceVinUniversity, Hanoi, Vietnam [email protected] Dung D. Pham MinhCollege of Engineering and Computer ScienceVinUniversity, Ha |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 13 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-14909