SleepVLM：基于视觉语言模型的可解释且规则驱动的睡眠分期
SleepVLM: A Rule-Grounded Vision-Language Model for Auditable Sleep Staging
AI总结：
提出SleepVLM，一种基于规则驱动的视觉语言模型，通过多通道PSG波形图像进行睡眠分期，并生成符合AASM评分标准的临床可读解释，在保持高准确率的同时提升可解释性。
摘要：
尽管自动睡眠分期已达到专家级准确率，但其临床采用因缺乏可审计的推理而受阻。我们提出了SleepVLM，一种基于规则驱动的视觉语言模型（VLM），它通过多通道多导睡眠图（PSG）波形图像进行睡眠分期，并基于美国睡眠医学学会（AASM）评分标准生成临床可读的理由。利用波形感知预训练和规则驱动的监督微调，SleepVLM在保留测试集（MASS-SS1）上实现了0.767的Cohen's kappa，在外部队列（ZUAMHCS）上实现了0.743，达到了最先进的性能。两位经过训练的睡眠技术专家的独立评估进一步验证了模型的推理质量，在两个数据集上，事实准确性、证据全面性和逻辑连贯性的平均得分在3.75-3.96之间（满分5分）。通过将竞争性性能与透明、基于规则的解释相结合，SleepVLM可以提高临床工作流程中自动睡眠分期的可信度和可审计性。为了促进可解释睡眠医学的进一步研究，我们发布了MASS-EX，一个新颖的专家注释数据集。
英文摘要：
Sleep staging is essential for sleep assessment and disorder diagnosis. In recent years, automatic sleep staging systems have achieved accuracy approaching that of human experts, but the black-box nature of their predictions hinders clinical adoption. Existing interpretability methods offer partial insight into model behavior, but their outputs still require expert reinterpretation and do not provide a direct basis for auditing individual predictions. To improve trustworthiness, we propose the task of auditable sleep staging. To solve this task, we present SleepVLM, a vision-language model that casts sleep staging as visual reasoning over rendered polysomnography (PSG) waveform images. For each epoch, SleepVLM outputs a stage together with the applicable American Academy of Sleep Medicine (AASM) rules and an auditable rationale. The model is trained using a two-stage framework: Waveform-Perceptual Pre-training followed by Rule-Grounded Supervised Fine-tuning over a mixture of fine-grained and coarse annotations. Experiments on four datasets show that SleepVLM outperforms state-of-the-art methods on average. An automated AASM-feature audit shows broad coverage of stage-defining evidence in the rationales, and independent experts validate their reasoning quality. To facilitate further research, we construct and release MASS-EX, an expert-annotated dataset for rule-grounded sleep staging with AASM rule annotations and expert-written rationales.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
