{"title":"A Survey of Few-Shot Learning for Biomedical Time Series.","authors":"Chenqi Li, Timothy Denison, Tingting Zhu","doi":"10.1109/RBME.2024.3492381","DOIUrl":null,"url":null,"abstract":"<p><p>Advancements in wearable sensor technologies and the digitization of medical records have contributed to the unprecedented ubiquity of biomedical time series data. Data-driven models have tremendous potential to assist clinical diagnosis and improve patient care by improving long-term monitoring capabilities, facilitating early disease detection and intervention, as well as promoting personalized healthcare delivery. However, accessing extensively labeled datasets to train data-hungry deep learning models encounters many barriers, such as long-tail distribution of rare diseases, cost of annotation, privacy and security concerns, data-sharing regulations, and ethical considerations. An emerging approach to overcome the scarcity of labeled data is to augment AI methods with human-like capabilities to leverage past experiences to learn new tasks with limited examples, called few-shot learning. This survey provides a comprehensive review and comparison of few-shot learning methods for biomedical time series applications. The clinical benefits and limitations of such methods are discussed in relation to traditional data-driven approaches. This paper aims to provide insights into the current landscape of few-shot learning for biomedical time series and its implications for future research and applications.</p>","PeriodicalId":39235,"journal":{"name":"IEEE Reviews in Biomedical Engineering","volume":"PP ","pages":""},"PeriodicalIF":17.2000,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Reviews in Biomedical Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1109/RBME.2024.3492381","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0
Abstract
Advancements in wearable sensor technologies and the digitization of medical records have contributed to the unprecedented ubiquity of biomedical time series data. Data-driven models have tremendous potential to assist clinical diagnosis and improve patient care by improving long-term monitoring capabilities, facilitating early disease detection and intervention, as well as promoting personalized healthcare delivery. However, accessing extensively labeled datasets to train data-hungry deep learning models encounters many barriers, such as long-tail distribution of rare diseases, cost of annotation, privacy and security concerns, data-sharing regulations, and ethical considerations. An emerging approach to overcome the scarcity of labeled data is to augment AI methods with human-like capabilities to leverage past experiences to learn new tasks with limited examples, called few-shot learning. This survey provides a comprehensive review and comparison of few-shot learning methods for biomedical time series applications. The clinical benefits and limitations of such methods are discussed in relation to traditional data-driven approaches. This paper aims to provide insights into the current landscape of few-shot learning for biomedical time series and its implications for future research and applications.
期刊介绍:
IEEE Reviews in Biomedical Engineering (RBME) serves as a platform to review the state-of-the-art and trends in the interdisciplinary field of biomedical engineering, which encompasses engineering, life sciences, and medicine. The journal aims to consolidate research and reviews for members of all IEEE societies interested in biomedical engineering. Recognizing the demand for comprehensive reviews among authors of various IEEE journals, RBME addresses this need by receiving, reviewing, and publishing scholarly works under one umbrella. It covers a broad spectrum, from historical to modern developments in biomedical engineering and the integration of technologies from various IEEE societies into the life sciences and medicine.