A one-dimensional convolutional neural network detected ganglionated plexi substrates from raw bipolar electrograms with an ROC-AUC of 0.870 and PR-AUC of 0.349 on external testing.
Does a one-dimensional convolutional neural network accurately detect ganglionated plexi from bipolar intracardiac electrograms compared to expert annotation?
A lightweight 1D CNN can detect ganglionated plexi substrates from bipolar intracardiac electrograms with high sensitivity, offering potential for real-time guidance in cardioneuroablation.
BACKGROUND: Precise localization of ganglionated plexi (GP) is critical for effective cardioneuroablation, yet current mapping relies on labour‑intensive stimulation and subjective electrogram (EGM) interpretation. Recent advancements in deep learning (DL) have shown the potential to automate and improve outcomes an atrial fibrillation by analyzing EGMs. We aimed to apply DL to raw bipolar EGMs in order to automate GP detection. METHODS: A total of 189 760 bipolar windows (18 left‑atrium and 15 right‑atrium maps, respectively) were collected from 18 patients. GP annotation was performed independently by two experienced electrophysiologists. Five atrial maps from three patients were withheld for external testing; the remaining 15 patients yielded 119 222 clean windows for model development (GP prevalence ≈ 3.5%). A lightweight one‑dimensional convolutional neural network (CNN) was implemented using PyTorch. Training used focal loss (α = 0.75, γ = 2.0) and class‑balanced sampling. Performance was assessed with ROC/PR curves, threshold sweeps and gradient‑weighted class activation mapping (GCAM) saliency mapping. RESULTS: On the validation set the model achieved 69.6% accuracy; GP precision, recall and F1‑score were 0.09, 0.85 and 0.17, respectively. External testing on 34 976 unseen windows produced ROC‑AUC = 0.870 and PR‑AUC = 0.349. A probability threshold of 0.70 captured 51% of reference GP sites while highlighting anatomically plausible "hot‑spots" (513/9 063 nodes). GCAM consistently focused on central waveform segments (indices140-160), aligning with fractionated autonomic signatures and reinforcing model interpretability. CONCLUSIONS: The proposed explainable one‑dimensional CNN detects GP substrates with high sensitivity despite pronounced class imbalance and generalizes to unseen atria. Its probability maps and saliency outputs provide intuitive visual guidance, supporting real‑time, physiology‑aware decision making in cardioneuroablation.
Güler et al. (2026) studied Atrial fibrillation / Ganglionated plexi localization (n=18). One-dimensional convolutional neural network (CNN) vs. Expert electrophysiologist annotation was evaluated on Model performance on external testing (ROC-AUC and PR-AUC). A one-dimensional convolutional neural network detected ganglionated plexi substrates from raw bipolar electrograms with an ROC-AUC of 0.870 and PR-AUC of 0.349 on external testing.