COMFORT:针对消费者医疗保健的基础模型持续微调框架

Chia-Hao Li, Niraj K. Jha
{"title":"COMFORT:针对消费者医疗保健的基础模型持续微调框架","authors":"Chia-Hao Li, Niraj K. Jha","doi":"arxiv-2409.09549","DOIUrl":null,"url":null,"abstract":"Wearable medical sensors (WMSs) are revolutionizing smart healthcare by\nenabling continuous, real-time monitoring of user physiological signals,\nespecially in the field of consumer healthcare. The integration of WMSs and\nmodern machine learning (ML) enables unprecedented solutions to efficient\nearly-stage disease detection. Despite the success of Transformers in various\nfields, their application to sensitive domains, such as smart healthcare,\nremains underexplored due to limited data accessibility and privacy concerns.\nTo bridge the gap between Transformer-based foundation models and WMS-based\ndisease detection, we propose COMFORT, a continual fine-tuning framework for\nfoundation models targeted at consumer healthcare. COMFORT introduces a novel\napproach for pre-training a Transformer-based foundation model on a large\ndataset of physiological signals exclusively collected from healthy individuals\nwith commercially available WMSs. We adopt a masked data modeling (MDM)\nobjective to pre-train this health foundation model. We then fine-tune the\nmodel using various parameter-efficient fine-tuning (PEFT) methods, such as\nlow-rank adaptation (LoRA) and its variants, to adapt it to various downstream\ndisease detection tasks that rely on WMS data. In addition, COMFORT continually\nstores the low-rank decomposition matrices obtained from the PEFT algorithms to\nconstruct a library for multi-disease detection. The COMFORT library enables\nscalable and memory-efficient disease detection on edge devices. Our\nexperimental results demonstrate that COMFORT achieves highly competitive\nperformance while reducing memory overhead by up to 52% relative to\nconventional methods. Thus, COMFORT paves the way for personalized and\nproactive solutions to efficient and effective early-stage disease detection\nfor consumer healthcare.","PeriodicalId":501541,"journal":{"name":"arXiv - CS - Human-Computer Interaction","volume":"6 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"COMFORT: A Continual Fine-Tuning Framework for Foundation Models Targeted at Consumer Healthcare\",\"authors\":\"Chia-Hao Li, Niraj K. Jha\",\"doi\":\"arxiv-2409.09549\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Wearable medical sensors (WMSs) are revolutionizing smart healthcare by\\nenabling continuous, real-time monitoring of user physiological signals,\\nespecially in the field of consumer healthcare. The integration of WMSs and\\nmodern machine learning (ML) enables unprecedented solutions to efficient\\nearly-stage disease detection. Despite the success of Transformers in various\\nfields, their application to sensitive domains, such as smart healthcare,\\nremains underexplored due to limited data accessibility and privacy concerns.\\nTo bridge the gap between Transformer-based foundation models and WMS-based\\ndisease detection, we propose COMFORT, a continual fine-tuning framework for\\nfoundation models targeted at consumer healthcare. COMFORT introduces a novel\\napproach for pre-training a Transformer-based foundation model on a large\\ndataset of physiological signals exclusively collected from healthy individuals\\nwith commercially available WMSs. We adopt a masked data modeling (MDM)\\nobjective to pre-train this health foundation model. We then fine-tune the\\nmodel using various parameter-efficient fine-tuning (PEFT) methods, such as\\nlow-rank adaptation (LoRA) and its variants, to adapt it to various downstream\\ndisease detection tasks that rely on WMS data. In addition, COMFORT continually\\nstores the low-rank decomposition matrices obtained from the PEFT algorithms to\\nconstruct a library for multi-disease detection. The COMFORT library enables\\nscalable and memory-efficient disease detection on edge devices. Our\\nexperimental results demonstrate that COMFORT achieves highly competitive\\nperformance while reducing memory overhead by up to 52% relative to\\nconventional methods. Thus, COMFORT paves the way for personalized and\\nproactive solutions to efficient and effective early-stage disease detection\\nfor consumer healthcare.\",\"PeriodicalId\":501541,\"journal\":{\"name\":\"arXiv - CS - Human-Computer Interaction\",\"volume\":\"6 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Human-Computer Interaction\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.09549\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Human-Computer Interaction","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.09549","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

可穿戴医疗传感器(WMS)通过实现对用户生理信号的连续、实时监测,正在彻底改变智能医疗保健,尤其是在消费者医疗保健领域。WMS 与现代机器学习(ML)的整合为高效的早期疾病检测提供了前所未有的解决方案。为了弥合基于 Transformer 的基础模型与基于 WMS 的疾病检测之间的差距,我们提出了 COMFORT,一个针对消费者医疗保健的基础模型的持续微调框架。COMFORT 引入了一种新颖的方法,用于在一个大型生理信号集上预训练基于变压器的基础模型,该生理信号集专门收集自使用市售 WMS 的健康个体。我们采用掩码数据建模(MDM)目标来预训练这一健康基础模型。然后,我们使用各种参数效率微调(PEFT)方法,如慢秩适应(LoRA)及其变体,对模型进行微调,使其适应依赖于 WMS 数据的各种下游疾病检测任务。此外,COMFORT 不断存储从 PEFT 算法中获得的低秩分解矩阵,以构建一个多疾病检测库。COMFORT 库可以在边缘设备上进行可扩展且内存效率高的疾病检测。实验结果表明,与传统方法相比,COMFORT 实现了极具竞争力的性能,同时将内存开销减少了 52%。因此,COMFORT 为消费者医疗保健领域高效、有效的早期疾病检测提供了个性化和前瞻性的解决方案。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
COMFORT: A Continual Fine-Tuning Framework for Foundation Models Targeted at Consumer Healthcare
Wearable medical sensors (WMSs) are revolutionizing smart healthcare by enabling continuous, real-time monitoring of user physiological signals, especially in the field of consumer healthcare. The integration of WMSs and modern machine learning (ML) enables unprecedented solutions to efficient early-stage disease detection. Despite the success of Transformers in various fields, their application to sensitive domains, such as smart healthcare, remains underexplored due to limited data accessibility and privacy concerns. To bridge the gap between Transformer-based foundation models and WMS-based disease detection, we propose COMFORT, a continual fine-tuning framework for foundation models targeted at consumer healthcare. COMFORT introduces a novel approach for pre-training a Transformer-based foundation model on a large dataset of physiological signals exclusively collected from healthy individuals with commercially available WMSs. We adopt a masked data modeling (MDM) objective to pre-train this health foundation model. We then fine-tune the model using various parameter-efficient fine-tuning (PEFT) methods, such as low-rank adaptation (LoRA) and its variants, to adapt it to various downstream disease detection tasks that rely on WMS data. In addition, COMFORT continually stores the low-rank decomposition matrices obtained from the PEFT algorithms to construct a library for multi-disease detection. The COMFORT library enables scalable and memory-efficient disease detection on edge devices. Our experimental results demonstrate that COMFORT achieves highly competitive performance while reducing memory overhead by up to 52% relative to conventional methods. Thus, COMFORT paves the way for personalized and proactive solutions to efficient and effective early-stage disease detection for consumer healthcare.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Equimetrics -- Applying HAR principles to equestrian activities AI paintings vs. Human Paintings? Deciphering Public Interactions and Perceptions towards AI-Generated Paintings on TikTok From Data Stories to Dialogues: A Randomised Controlled Trial of Generative AI Agents and Data Storytelling in Enhancing Data Visualisation Comprehension Exploring Gaze Pattern in Autistic Children: Clustering, Visualization, and Prediction Revealing the Challenge of Detecting Character Knowledge Errors in LLM Role-Playing
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1