Isolated REM sleep behavior disorder (iRBD) is a key prodromal marker of Parkinson's disease (PD), and video-polysomnography (vPSG) remains the diagnostic gold standard. However, manual sleep staging is particularly challenging in neurodegenerative diseases due to EEG abnormalities and fragmented sleep, making PSG assessments a bottleneck for deploying new RBD screening technologies at scale. We adapted U-Sleep, a deep neural network, for generalizable sleep staging in PD and iRBD. A pretrained U-Sleep model, based on a large, multisite non-neurodegenerative dataset (PUB; 19,236 PSGs across 12 sites), was fine-tuned on research datasets from two centers (Lundbeck Foundation Parkinson's Disease Research Center (PACE) and the Cologne-Bonn Cohort (CBC); 112 PD, 138 iRBD, 89 age-matched controls. The resulting model was evaluated on an independent dataset from the Danish Center for Sleep Medicine (DCSM; 81 PD, 36 iRBD, 87 sleep-clinic controls). A subset of PSGs with low agreement between the human rater and the model (Cohen's $κ$ < 0.6) was re-scored by a second blinded human rater to identify sources of disagreement. Finally, we applied confidence-based thresholds to optimize REM sleep staging. The pretrained model achieved mean $κ$ = 0.81 in PUB, but $κ$ = 0.66 when applied directly to PACE/CBC. By fine-tuning the model, we developed a generalized model with $κ$ = 0.74 on PACE/CBC (p < 0.001 vs. the pretrained model). In DCSM, mean and median $κ$ increased from 0.60 to 0.64 (p < 0.001) and 0.64 to 0.69 (p < 0.001), respectively. In the interrater study, PSGs with low agreement between the model and the initial scorer showed similarly low agreement between human scorers. Applying a confidence threshold increased the proportion of correctly identified REM sleep epochs from 85% to 95.5%, while preserving sufficient (> 5 min) REM sleep for 95% of subjects.

翻译：孤立性快速眼动睡眠行为障碍（iRBD）是帕金森病（PD）的关键前驱标志物，而视频多导睡眠监测（vPSG）仍是诊断的金标准。然而，在神经退行性疾病中，由于脑电图异常和睡眠片段化，人工睡眠分期尤为困难，这使得PSG评估成为大规模部署新型RBD筛查技术的瓶颈。我们针对PD和iRBD，对深度神经网络U-Sleep进行了适应性调整，以实现可泛化的睡眠分期。一个基于大型、多中心非神经退行性疾病数据集（PUB；来自12个中心的19,236份PSG）预训练的U-Sleep模型，在两个研究中心（隆德贝克基金会帕金森病研究中心（PACE）和科隆-波恩队列（CBC）；112例PD，138例iRBD，89例年龄匹配对照）的研究数据集上进行了微调。所得模型在一个来自丹麦睡眠医学中心（DCSM；81例PD，36例iRBD，87例睡眠门诊对照）的独立数据集上进行了评估。对于人类评分者与模型之间一致性较低（Cohen's $κ$ < 0.6）的PSG子集，由第二位盲法人类评分者重新评分，以识别不一致的来源。最后，我们应用基于置信度的阈值来优化快速眼动睡眠分期。预训练模型在PUB中平均$κ$ = 0.81，但直接应用于PACE/CBC时$κ$ = 0.66。通过对模型进行微调，我们开发了一个在PACE/CBC上$κ$ = 0.74的泛化模型（与预训练模型相比，p < 0.001）。在DCSM中，平均和中位$κ$分别从0.60增加到0.64（p < 0.001）和从0.64增加到0.69（p < 0.001）。在评分者间研究中，模型与初始评分者之间一致性较低的PSG，在人类评分者之间也表现出类似较低的一致性。应用置信度阈值后，正确识别的快速眼动睡眠时段比例从85%提高到95.5%，同时为95%的受试者保留了足够（> 5分钟）的快速眼动睡眠。