Tracking biomedical articles along the translational continuum: a measure based on biomedical knowledge...

Area of Science:

• Biomedical Informatics
• Translational Medicine
• Bibliometrics

Background:

• Evaluating translational medicine program performance requires robust metrics.
• Existing indicators lack consensus for article-level translational feature assessment.
• A need exists for a standardized, scalable measure of biomedical research translation.

Purpose of the Study:

• To develop and validate a novel measure, Translational Progression (TP), for tracking biomedical research along the translational continuum.
• To leverage semantic representations (bio-entity2vec, bio-doc2vec) for quantifying translational features.
• To assess the utility of TP in real-time monitoring and decision-making for translational research.

Main Methods:

• Trained semantic representations (bio-entity2vec, bio-doc2vec) on over 30 million PubMed articles.
• Developed the Translational Progression (TP) measure based on these semantic vectors.
• Validated TP against clinical trial phase identification and ACH classification for consistency.

Main Results:

• TP demonstrated excellent consistency with established indicators for clinical trial phase and ACH classification.
• TP effectively tracks the degree of translation dynamically and in real-time.
• Analysis revealed significant findings regarding the distribution, temporal trends, and topic-specific nature of translational progression.

Conclusions:

• Translational Progression (TP) provides a dynamic, interpretable, and scalable method for assessing biomedical research translation.
• TP overcomes limitations of manual labeling and PubMed indexing, suitable for big data analysis.
• The TP measure can inform policy, guide research investment, and be adapted for evaluating scientific application value in other fields.