What question did this study set out to answer?

The review aims to explore the current state of few-shot learning methods for medical natural language processing.

July 23, 2023Journal of Biomedical Informatics105 citationsOpen Access

Few-shot learning for medical text: A review of advances, trends, and opportunities

Key Points

The review aims to explore the current state of few-shot learning methods for medical natural language processing.
Conducted a systematic search of articles published from January 2016 to October 2022 across multiple databases.
Included articles focused on few-shot learning and medical text, leading to a final inclusion of fifty-one articles.
Analyzed data sources, target tasks, training set sizes, methods, and evaluation metrics.
Most articles were published after 2018, with 82% published since 2020.
Concept extraction and named entity recognition were the most common tasks, occurring in 41% of articles.
Many articles used FSL with attention mechanisms and sought to augment small datasets with prior knowledge.

Abstract

BACKGROUND: Few-shot learning (FSL) is a class of machine learning methods that require small numbers of labeled instances for training. With many medical topics having limited annotated text-based data in practical settings, FSL-based natural language processing (NLP) holds substantial promise. We aimed to conduct a review to explore the current state of FSL methods for medical NLP. METHODS: We searched for articles published between January 2016 and October 2022 using PubMed/Medline, Embase, ACL Anthology, and IEEE Xplore Digital Library. We also searched the preprint servers (e.g., arXiv, medRxiv, and bioRxiv) via Google Scholar to identify the latest relevant methods. We included all articles that involved FSL and any form of medical text. We abstracted articles based on the data source, target task, training set size, primary method(s)/approach(es), and evaluation metric(s). RESULTS: Fifty-one articles met our inclusion criteria-all published after 2018, and most since 2020 (42/51; 82%). Concept extraction/named entity recognition was the most frequently addressed task (21/51; 41%), followed by text classification (16/51; 31%). Thirty-two (61%) articles reconstructed existing datasets to fit few-shot scenarios, and MIMIC-III was the most frequently used dataset (10/51; 20%). 77% of the articles attempted to incorporate prior knowledge to augment the small datasets available for training. Common methods included FSL with attention mechanisms (20/51; 39%), prototypical networks (11/51; 22%), meta-learning (7/51; 14%), and prompt-based learning methods, the latter being particularly popular since 2021. Benchmarking experiments demonstrated relative underperformance of FSL methods on biomedical NLP tasks. CONCLUSION: Despite the potential for FSL in biomedical NLP, progress has been limited. This may be attributed to the rarity of specialized data, lack of standardized evaluation criteria, and the underperformance of FSL methods on biomedical topics. The creation of publicly-available specialized datasets for biomedical FSL may aid method development by facilitating comparative analyses.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper