arXiv cs.AI· Walentin Palanchuk·· 12 小时前AI 评分31
用 ECG 预测实验室检验值：一项生理与技术综合分析
ECG for laboratory values:A comprehensive physiological and technical analysis
AI 导读
一项基于 MIMIC-IV 的多模态深度学习研究用 ECG 与患者元数据预测 NT-proBNP，12 导联配置表现稳定优于单导联 II：回归 MAE 为 514.27 pg/mL 对 534.98 pg/mL，二分类 AUROC 为 0.876 对 0.849，三分类宏平均 AUROC 为 0.752 对 0.738。
正文 · 原文
Abstract:Laboratory measurements are essential for clinical decision making but typically require invasive blood sampling. Recent research suggests that ECG recordings may contain information beyond cardiac rhythm and provide information related to laboratory values. This thesis investigates NT-proBNP prediction from ECG recordings and patient metadata using a multimodal deep learning framework, while considering the physiological plausibility of ECG-based prediction. It also compares a 1-lead configuration using lead II with the corresponding 12-lead configuration. Additional experiments on glucose, hemoglobin, and potassium are presented in the appendix.
Regression, binary classification, and three-class classification experiments were conducted using MIMIC-IV, with the three-class task based on clinically established thresholds. The 12-lead configuration consistently achieved better performance. Regression reached a mean absolute error of 534.98 pg/mL for 1-lead and 514.27 pg/mL for 12-lead. Binary classification achieved AUROCs of 0.849 and 0.876, respectively, while three-class classification achieved macro-averaged AUROCs of 0.738 and 0.752.
The results indicate that precise NT-proBNP regression remains challenging, whereas broader concentration ranges can be classified more effectively. The findings are consistent with the physiological hypothesis that ECG characteristics and NT-proBNP concentrations reflect shared cardiovascular processes. ECG-based NT-proBNP prediction may therefore provide complementary information, particularly for classifying broader concentration ranges. However, laboratory measurements remain necessary, and validation on independent datasets is required.
| Comments: | 78 pages, 17 figures, source code under this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.28507 [cs.LG] |
| (or arXiv:2610.28507v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28507 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Walentin Palanchuk [view email]
[v1]
Tue, 20 Oct 2026 16:46:26 UTC (2,065 KB)
来源：arXiv cs.AI · arxiv.org