A survey of few-shot learning for biomedical time series

Journal article

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.

Authors: Li, C; Denison, T; Zhu, T

• Publication status: In press
• Peer review status: Peer reviewed
• Publisher: IEEE
• Journal: IEEE Reviews in Biomedical Engineering
• Volume: 18
• Pages: 192-210
• Place of publication: United States
• Publication date: 2024-11-06
• Acceptance date: 2024-10-29
• DOI: 10.1109/rbme.2024.3492381
• EISSN: 1941-1189
• ISSN: 1937-3333
• Pmid: 39504299
• Language: English
• Keywords: clinical research; bioengineering; behavioral and social science; networking and information technology r&d (nitrd); generic health relevance; machine learning and artificial intelligence