ECG-derived Poincaré plot graphical metrics for differentiating axillary lymph node metastasis in breast cancer
Frontiers in Physiology · 20.05.2026 · 17
DOI: 10.3389/fphys.2026.1795471 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Physiology, online erschienen am 20.05.2026. Verfasst von Nannan Fang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Anhui Key Laboratory of Digital Medicine and Intelligent Health, School of Medical Imaging, Bengbu Medical University, Bengbu, China. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Bengbu Medical College, Anhui Provincial Department of Education.
Abstract (Originalsprache)
Background
Axillary lymph node metastasis (ALNM) status serves as a cornerstone for clinical decision-making in personalized breast cancer (BC) management. Although autonomic nervous system (ANS) dysfunction correlates with tumor progression, its applicability for capturing BC-ALNM-specific ANS signatures remains exploratory. This proof-of-concept study aimed to investigate whether Poincaré plot analysis of short-duration electrocardiograms could potentially identify ANS signals associated with ALNM more effectively than conventional heart rate variability (HRV) metrics.
Methods
Resting electrocardiograph data were collected from 218 pathologically confirmed BC patients at the First Affiliated Hospital of Bengbu Medical University (142 ALNM-positive, 76 ALNM-negative). Conventional HRV metrics and Poincaré plot-derived spatial metrics were compared using raw and detrended data. Beyond univariable comparisons evaluated by Cohen’s d effect sizes with 95% confidence intervals (CIs), multivariable Analysis of Covariance (ANCOVA) adjusting for clinicopathological confounders were employed to ensure model robustness. Discriminative performance was evaluated using Receiver Operating Characteristic (ROC) analysis. To assess the potential for overfitting, a 10-fold cross-validation procedure was implemented for the ROC analysis.
Results
Conventional metrics showed no significant intergroup differences or predictive value (AUC = 0.532). Univariable analysis revealed differences in the detrended Area Index (AI) ( p = 0.006, Cohen’s d = 0.40, 95% CI: 0.12-0.68), and raw grid distribution metrics (GDR, GDE) showed significant differences between groups. However, after strict multivariable adjustment, only the detrended AI remained a significant independent predictor (adjusted p = 0.024). The original AUC of the detrended AI was 0.612, and the 10-fold cross-validated AUC was 0.607, indicating robustness against overfitting.
Conclusions
In this exploratory study, Poincaré plot analysis, specifically the detrended AI, suggests a potential methodological advantage for characterizing ALNM-associated autonomic spatial asymmetry independent of general tumor burden. However, given the modest discriminative ability and single-center design, these findings must be interpreted with caution. These non-invasive metrics may offer a novel pathophysiological perspective but require future large-scale, multi-center prospective studies to overcome current limitations and validate their clinical utility.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Anhui Key Laboratory of Digital Medicine and Intelligent Health, School of Medical Imaging, Bengbu Medical University, Bengbu, China.
- Department of Radiology, The First Affiliated Hospital of Bengbu Medical University, Bengbu, Anhui, China.
Förderung
- Bengbu Medical College
- Anhui Provincial Department of Education
Zitieren
APA
Fang, N., Zhang, A., Song, Z., Xie, Z., Shi, B., & Wang, Z. (2026). ECG-derived Poincaré plot graphical metrics for differentiating axillary lymph node metastasis in breast cancer. Frontiers in Physiology, 17. https://doi.org/10.3389/fphys.2026.1795471
Vancouver
Fang N, Zhang A, Song Z, Xie Z, Shi B, Wang Z. ECG-derived Poincaré plot graphical metrics for differentiating axillary lymph node metastasis in breast cancer. Frontiers in Physiology. 2026;17. doi:10.3389/fphys.2026.1795471
BibTeX
@article{Fang2026,
author = {Fang, Nannan and Zhang, Aoqi and Song, Zhijing and Xie, Zongyu and Shi, Bo and Wang, Zhiwen},
title = {ECG-derived Poincaré plot graphical metrics for differentiating axillary lymph node metastasis in breast cancer},
journal = {Frontiers in Physiology},
year = {2026},
volume = {17},
doi = {10.3389/fphys.2026.1795471}
}
EUVIMED