Key result
Combining ECG and PPG via late fusion raises AF detection AUC by ~0.02 over ECG alone.
Why the study?
Cardiac monitors rarely record every signal, and the value of adding signals through decision-level late fusion of ECG, PCG, and PPG for cardiovascular disease detection is seldom evaluated.
Does calibrated, modality-agnostic late fusion of ECG, PCG, and PPG signals improve cardiovascular disease detection discrimination and scoring coverage compared to single-modality or complete-case models?
Does calibrated, modality-agnostic late fusion of ECG, PCG, and PPG signals improve cardiovascular disease detection discrimination and scoring coverage compared to single-modality or complete-case models?
Calibrated, modality-agnostic late fusion enables cardiovascular disease detection across all records regardless of missing or low-quality signals (such as PPG failing quality checks 3.7 times more often during atrial fibrillation), improving discrimination over ECG alone when complementary signal information exists.
Answers are grounded in this paper and cite their sources.
Calibrated late fusion may boost AF detection coverage and AUC; leaves open prospective outcome validation before clinical use.
Cardiac monitors rarely record every signal, and adding one is seldom evaluated. We asked when decision-level (late) fusion of the electrocardiogram (ECG), phonocardiogram (PCG) and photoplethysmogram (PPG) improves cardiovascular disease detection, and why, using one model for any signal combination. One expert per signal was trained in nested, subject-grouped cross-validation, Platt-calibrated and combined by seven rules, each fitted once and applied to any available signals. Three studies: abnormal heart sounds (PhysioNet/CinC 2016, ECG+PCG, 329 recordings), atrial fibrillation (AF; MIMIC-III-Ext-PPG, ECG+PPG, 40,792 segments, 5,642 patients) and a synthetic tri-modal (ECG+PCG+PPG) benchmark. PPG failed quality checks 3.7 times more often in AF, so complete-case fusion could not score 43.9% of AF segments; the modality-agnostic model scored every segment, raising the area under the receiver operating characteristic curve (AUC) over ECG alone from 0.923 to 0.938 (+0.015, 95% confidence interval 0.012--0.018), but only with calibrated experts (uncalibrated: 0.847). With both signals, PPG raised AUC from 0.910 to 0.931 (+0.021, 0.017--0.026). PCG added nothing (p\,=\,1.00), consistent with published gains (0.004--0.023). Synthetic data favoured all three signals (AUC 0.984; best pair 0.976). Soft voting nearly matched the best rule; hard voting lost because two votes give three scores. Correlated expert errors, not the combiner, bounded the gain: the best rule nearly matched an oracle combination. Expected Gradients explained the ECG and PCG experts faithfully. Calibrated, modality-agnostic late fusion scores every record from its usable signals; its gain is bounded by expert redundancy, measurable before adding sensors.
No takes yet. Share an insight, caveat, or question.
Alam et al. (2026) studied Cardiovascular disease detection (atrial fibrillation and abnormal heart sounds) (n=5,642). Calibrated modality-agnostic decision-level (late) fusion of ECG, PCG, and PPG signals vs. Single-modality expert models (ECG alone) or complete-case multimodal fusion was evaluated on Area under the receiver operating characteristic curve (AUC) for atrial fibrillation detection across all 40,792 segments in MIMIC-III-Ext-PPG (AUC difference +0.015, 95% CI 0.012-0.018). Calibrated modality-agnostic late fusion of ECG and PPG scored 100% of segments and raised atrial fibrillation detection AUC over ECG alone from 0.923 to 0.938 (+0.015, 95% CI 0.012-0.018).
Synapse has enriched 10 related papers. Consider them for comparative context: