Emerging trend prediction in biomedical literature

Area of Science:

• Biomedical informatics
• Natural Language Processing
• Knowledge Discovery

Background:

• Identifying emerging trends in biomedical research is crucial for scientific advancement.
• Existing methods may not capture the dynamic evolution of biomedical knowledge.
• The Medical Subject Headings (MeSH) ontology represents a structured vocabulary for indexing biomedical literature.

Purpose of the Study:

• To develop a method for predicting emerging trends in the biomedical domain using textual data.
• To anticipate the integration of new terms into the MeSH ontology as a proxy for emerging knowledge.
• To create and utilize a dataset for evaluating emerging trend prediction algorithms.

Main Methods:

• Analysis of textual data from biomedical literature, specifically PubMed abstracts.
• Developing algorithms to predict the inclusion of new terms into the MeSH ontology.
• Dataset preparation using PubMed abstracts and associated MeSH terms for evaluation.

Main Results:

• Demonstrated the feasibility of early prediction of emerging biomedical trends.
• Showcased a method for transforming information into knowledge by tracking MeSH ontology updates.
• Successfully prepared a dataset for evaluating trend prediction models.

Conclusions:

• Early prediction of emerging trends in the biomedical field is achievable.
• Monitoring MeSH ontology changes can serve as an indicator of emerging biomedical knowledge.
• The proposed approach offers a novel way to track and anticipate scientific advancements.