发表机构
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出面向女性健康的专用可穿戴基础模型FemWear，通过参数高效方式改造预训练主干，在女性健康相关指标上取得性能提升，但未确立普遍优势与临床有效性。
AI 中文摘要
通用可穿戴设备基础模型在广泛的传感器数据流和人群上进行预训练，但并非围绕女性健康任务设计。本文介绍FemWear，一种专用可穿戴设备基础模型，它以参数高效的方式重新利用预训练的多模态可穿戴设备主干。FemWear保留了补丁投影和Transformer编码器，通过低秩残差适配器和因果任务系列头训练239236个参数（占2154万参数编码器的1.11%）。它学习一个共享的纵向表征，涵盖月经、症状、情感、睡眠/恢复、自主神经、活动以及与妊娠相关的结局。我们评估了6个队列，包含63个可比的主要指标，其中33个来自女性健康队列，同时保留了32任务的OpenMHC能力保留基准。在固定参与者划分的3个随机种子上，FemWear使周期阶段的宏F1提升8.15%，并将痉挛、情绪症状和睡眠问题的平均绝对误差分别降低9.32%、5.80%和9.43%。在更严格的42名参与者的嵌套留一参与者审计中，24小时发作、72小时发作和痉挛分别保持了2.87%、6.35%和2.19%的正向变化；阶段、情绪和睡眠的结果为中性或负面，且没有任何指标具有严格正向的校正置信区间。容量匹配实验的表现优于最新的多层感知机，但未超过共享GRU或多门混合专家基线。仅训练校准将发作的预期校准误差降低了84.2%至88.2%，且无时间嵌套违规。FemWear为女性健康研究实现了针对性迁移和连贯的概率输出，但未确立普遍的性能优势或临床有效性。
英文摘要
General-purpose wearable foundation models are pretrained on broad sensor streams and populations, but their representations are not organized around women's health. FemWear is a women's wearable foundation model, obtained by parameter-efficiently repurposing a pretrained general multimodal wearable backbone into a specialized representation for women's health. It keeps the pretrained patch projection and Transformer encoder frozen and trains 239,236 encoder parameters - 1.11% of a 21.54M-parameter encoder - through low-rank residual adapters and causal task-family heads, producing one shared longitudinal representation for menstrual, symptom, affective, sleep/recovery, autonomic, activity, and pregnancy outcomes. We evaluate six cohorts with 63 comparable primary metrics, 33 from women's-health cohorts, while retaining the 32-task OpenMHC ability-retention benchmark. On a fixed participant split over three seeds, FemWear improved cycle-phase macro-F1 by 8.15% and reduced mean absolute error for cramps, mood symptoms, and sleep problems by 9.32%, 5.80%, and 9.43%; 24-hour onset AUPRC decreased by 3.40%. A stricter 42-participant nested leave-one-participant-out audit retained positive changes for 24-hour onset (+2.87%), 72-hour onset (+6.35%), and cramps (+2.19%), while phase, mood, and sleep changes were neutral or negative and no endpoint had a strictly positive corrected confidence interval. Capacity-matched experiments beat a latest-day multilayer perceptron but not shared-GRU or multi-gate mixture-of-experts baselines. Train-only calibration reduced onset expected calibration error by 84.2-88.2% with zero temporal-nesting violations. FemWear is therefore a women's wearable foundation model: a reproducible, parameter-efficient specialization delivering targeted transfer across women's-health tasks and coherent probability outputs.
Comments11 pages, 4 figures
