Personalized Scorer Modeling: A Learning-Based Framework for Deriving Robust Sleep Stage Labels from Multiple Experts
Summary
This paper presents a learning-based framework that models individual scorer behavior via confusion matrices to derive more robust sleep stage labels from multiple experts, improving accuracy on DOD-H and DOD-O datasets.
View Cached Full Text
Cached at: 08/14/26, 09:29 AM
# Personalized Scorer Modeling: A Learning-Based Framework for Deriving Robust Sleep Stage Labels from Multiple Experts Source: [https://arxiv.org/abs/2608.12446](https://arxiv.org/abs/2608.12446) [View PDF](https://arxiv.org/pdf/2608.12446) > Abstract:Sleep stage classification is important for the diagnosis and management of sleep disorders, yet most automatic staging studies evaluate models against a single reference hypnogram despite known inter\-scorer variability\. This study investigates whether multi\-scored datasets can be used to construct more reliable reference labels from the collective behavior of multiple experts\. We use the publicly available DOD\-H and DOD\-O datasets\. EEG \(C3\-M2\) and chin EMG signals were segmented into 30\-s epochs, and 30 features were extracted from each modality, yielding 60 features for EEG\+EMG\. We propose a learning\-based hypnogram \(LBH\) that models the stage\-specific behavior of each scorer using confusion matrices derived from machine\-learning models\. After column normalization, these matrices estimate the probability of each true sleep stage given each scorer's label; probabilities are aggregated across scorers to assign the final label for each epoch\. LBH was evaluated with random forest, support vector machine, and multilayer perceptron classifiers under EEG\-only and EEG\+EMG settings, and compared with the dataset hypnogram \(DH\) and best\-scorer hypnogram \(BSH\)\. LBH consistently improved overall performance\. The best results were obtained with random forest and EEG\+EMG, reaching 86\.07% accuracy, 85\.46% precision, and 85\.29% F1\-score on DOD\-H, and 86\.04% accuracy, 85\.21% precision, and 84\.70% F1\-score on DOD\-O\. These findings suggest that personalized scorer modeling can improve reference hypnogram construction without discarding information from individual experts\. ## Submission history From: AmirHossein Eshghi \[[view email](https://arxiv.org/show-email/74af7fc0/2608.12446)\] **\[v1\]**Wed, 12 Aug 2026 17:28:43 UTC \(1,464 KB\)
Similar Articles
Staging by the Book: Automatic Sleep Stage Classification Using Scoring Rules
This paper presents a deterministic, rule-based sleep staging method that explicitly implements the American Academy of Sleep Medicine (AASM) scoring rules, providing epoch-level natural language explanations. It achieves 60.5% epoch-level agreement with a majority-vote consensus on 50 polysomnography recordings, offering transparency as a complement to opaque deep learning models.
A Conflict-aware Evidential Framework for Reliable Sleep Stage Classification
ConfSleepNet is a conflict-aware evidential framework for reliable sleep stage classification using multi-modal data. It introduces hybrid category structures and a conflict-aware aggregation method to resolve inter-view conflicts, demonstrating effectiveness on sleep staging tasks.
Beyond AHI: An Interpretable Causal-Discovery-Guided Framework for Sleep Recovery in Connected Health
This paper proposes an interpretable causal-discovery-guided framework for deriving a Sleep Recovery Score (SRS) from multimodal polysomnography data, demonstrating up to 2.5× stronger alignment with perceived recovery than the traditional Apnea–Hypopnea Index (AHI), with potential applications in connected health.
STDA-Net: Spectrogram-Based Domain Adaptation for cross-dataset Sleep Stage Classification
This paper introduces STDA-Net, a domain adaptation framework for cross-dataset sleep stage classification using 2D spectrograms and adversarial learning. It demonstrates improved accuracy and stability over existing 1D EEG baseline methods on public datasets.
Omni-Sleep: A Sleep Foundation Model via Hierarchical Contrastive Learning of CNS--ANS Dynamic
Omni-Sleep is a sleep foundation model that uses hierarchical contrastive learning to capture CNS-ANS dynamics from multimodal polysomnography signals, outperforming strong baselines on sleep staging and multi-disease classification.