Apnea Burden-Guided Framework: Enhancing Out-of-Distribution Generalization in PPG-Based Sleep Apnea Characterization

Sleep apnea is a common sleep-related breathing disorder associated with substantial cardiovascular and metabolic risk. Although overnight polysomnography remains the reference standard for diagnosis, its complexity and cost limit its suitability for long-term preventive monitoring at home. In wearable systems, arterial blood oxygen saturation (SpO2) is commonly used as the main predictor, whereas additional morphological features of the photoplethysmographic (PPG) pulse wave are usually underexplored. This study proposes a novel apnea burden prediction-based framework for sleep apnea severity assessment and investigates the influence of PPG features on model performance and out-of-distribution (OOD) generalization. The proposed framework first predicts apnea burden as a continuous measure, which is subsequently converted into the clinically relevant apnea-hypopnea index for subject-level classification into four severity groups in both in-distribution (ID) and OOD data. Three artificial neural network architectures were evaluated, and the performance metrics were averaged over five independent runs with different fixed random seeds. During OOD testing, the combination of PPG features and SpO2 led to increases of approximately 15.72% in macro-sensitivity, 9.22% in macro-accuracy, 16.01% in macro-F1-score, 11.08% in Cohen's kappa, and 13.22% in Matthews correlation coefficient, compared with using SpO2 alone. The low-complexity convolutional-recurrent models achieved the highest overall performance. The results indicated that the proposed apnea burden-guided framework, combined with PPG features and SpO2, improves sleep apnea characterization while showing encouraging OOD performance on an independent external dataset. Moreover, simpler hybrid architectures demonstrated strong potential for robust home-based preventive monitoring.

Effect of Photoplethysmogram Artifacts on Uncertainty in Atrial Fibrillation Detection

Detection of atrial fibrillation (AF) from photoplethysmogram (PPG) is highly sensitive to artifacts, yet their effect on uncertainty of different AF detectors remains poorly understood. This work aims to quantify how different PPG artifact types affect the uncertainty of AF detectors. Two machine learning approaches to AF detection were explored: one using 25-s PPG signals as input ($\mathcal{D}_{r}$) and another using AF-related rhythm irregularity features ($\mathcal{D}_{f}$). The detectors were trained on wrist PPG signals acquired during cardiac rehabilitation and then systematically evaluated on 260,000 PPG signals containing controlled artifact types and durations. Uncertainty was quantified using a threshold-based error rate, conformal prediction, and Monte Carlo dropout. Using artifact-free PPG signals, $\mathcal{D}_{f}$ outperforms $\mathcal{D}_{r}$ with sensitivity/specificity of 0.94/0.94 versus 0.92/0.86. Relative to artifact-free performance, sensitivity/specificity drops by 0.52/0.03, 0.25/0.03, 0.16/0.02, and 0.07/0.02 using $\mathcal{D}_{r}$ for 12-s artifacts of device displacement, forearm motion, hand motion, and poor contact respectively. For $\mathcal{D}_{f}$, the respective drops are 0.31/0.07, 0.19/0.11, 0.14/0.15, and 0.13/0.21 for the same artifacts. $\mathcal{D}_{f}$ is more robust to short artifacts but exhibits increasing uncertainty with longer artifact durations, whereas $\mathcal{D}_{r}$ shows an abrupt performance drop when artifacts occur but is less sensitive to artifact duration. Applying conformal prediction with 90\% coverage reduces the false-positive rate by up to 12\% for $\mathcal{D}_{r}$ and up to 64\% for $\mathcal{D}_{f}$. Artifact type and duration have detector-specific effects on AF detection uncertainty. Device displacement causes the largest increase in uncertainty.

Benchmark Problems and Benchmark Datasets for the evaluation of Machine and Deep Learning methods on Photoplethysmography signals: the D4 report from the QUMPHY project

This report is part of the Qumphy project (22HLT01 Qumphy) that is funded by the European Union and is dedicated to the development of measures to quantify the uncertainties associated with Machine Learning algorithms applied to medical problems, in particular the analysis and processing of Photoplethysmography (PPG) signals. In this report, a list of six medical problems that are related to PPG signals and serve as Benchmark Problems is given. Suitable Benchmark datasets and their usage are described also.

Obstructive Sleep Apnea Characterization: A Multimodal Cross-Recurrence-Based Approach for Investigating Atrial Fibrillation

Obstructive sleep apnea (OSA) is believed to contribute significantly to atrial fibrillation (AF) development in certain patients. Recent studies indicate a rising risk of AF with increasing OSA severity. However, the commonly used apnea-hypopnea index in clinical practice may not adequately account for the potential cardiovascular risks associated with OSA. (1) Objective: to propose and explore a novel method for assessing OSA severity considering potential connection to cardiac arrhythmias. (2) Method: the approach utilizes cross-recurrence features to characterize OSA and AF by considering the relationships among oxygen desaturation, pulse arrival time, and heart-beat intervals. Multinomial logistic regression models were trained to predict four levels of OSA severity and four groups related to heart rhythm issues. The rank biserial correlation coefficient, rrb, was used to estimate effect size for statistical analysis. The investigation was conducted using the MESA database, which includes polysomnography data from 2055 subjects. (3) Results: a derived cross-recurrence-based index showed a significant association with a higher OSA severity (p < 0.01) and the presence of AF (p < 0.01). Additionally, the proposed index had a significantly larger effect, rrb, than the conventional apnea-hypopnea index in differentiating increasingly severe heart rhythm issue groups: 0.14 > 0.06, 0.33 > 0.10, and 0.41 > 0.07. (4) Significance: the proposed method holds relevance as a supplementary diagnostic tool for assessing the authentic state of sleep apnea in clinical practice.

Machine-learning for photoplethysmography analysis: Benchmarking feature, image, and signal-based approaches

Photoplethysmography (PPG) is a widely used non-invasive physiological sensing technique, suitable for various clinical applications. Such clinical applications are increasingly supported by machine learning methods, raising the question of the most appropriate input representation and model choice. Comprehensive comparisons, in particular across different input representations, are scarce. We address this gap in the research landscape by a comprehensive benchmarking study covering three kinds of input representations, interpretable features, image representations and raw waveforms, across prototypical regression and classification use cases: blood pressure and atrial fibrillation prediction. In both cases, the best results are achieved by deep neural networks operating on raw time series as input representations. Within this model class, best results are achieved by modern convolutional neural networks (CNNs). but depending on the task setup, shallow CNNs are often also very competitive. We envision that these results will be insightful for researchers to guide their choice on machine learning tasks for PPG data, even beyond the use cases presented in this work.