PubMed

**Group 1: PubMed Overview**

– PubMed is a free database maintained by the United States National Library of Medicine (NLM) at the National Institutes of Health.
– It primarily includes the MEDLINE database of references and abstracts on life sciences and biomedical topics.
– PubMed was first released in January 1996, offering private, free, home- and office-based MEDLINE searching.
– As of May 23, 2023, PubMed has more than 35 million citations and abstracts dating back to 1966.
– Over the last 10 years, nearly one million new records were added to PubMed each year.

**Group 2: PubMed Search and Navigation**

– A new PubMed interface was launched in October 2009, with features for quick, Google-like search formulations.
– Results on PubMed can be sorted by various parameters such as Most Recent, Best Match, Publication Date, First Author, Last Author, Journal, or Title.
– Simple searches on PubMed can be enhanced by using relevant MeSH terms, synonyms, and Boolean operators.
– Understanding the core component MEDLINE and MeSH controlled vocabulary is essential for optimal searches.
– PubMed automatically links to MeSH terms and subheadings for improved search accuracy.

**Group 3: PubMed Features and Functions**

– My NCBI provides tools for saving searches, filtering results, setting up automatic updates, and configuring display formats.
– LinkOut is an NLM facility allowing access to full-text local journal holdings from approximately 3,200 sites.
– PubMed Commons, a platform for authors to comment on articles, was tested in pilot mode in 2013 and discontinued in February 2018.
– askMEDLINE is a free-text, natural language query tool for MEDLINE/PubMed, suitable for handheld devices.
– PubMed identifier is a unique integer value assigned to each PubMed record, distinct from a PMCID, and can be used to locate references.

**Group 4: PubMed Data Usage and Mining**

– Methods like Matlab, Python, or R are used to mine PubMed data, which is also employed for training biomedical language models.
– Various models like PubMedGPT and BiomedCLIP-PubMedBERT have been developed for AI in medical research using PubMed data.
– PubMed data can be mirrored locally using tools like MEDOC and is utilized in open data datasets and data analysis tools for libraries.

**Group 5: PubMed Related Tools and Developments**

– Tools like MedlineRanker, MiSearch, Anne OTate, and ClusterMed have been developed for flexible ranking, adaptive search, summarization, and clustering of biomedical literature.
– Other tools like EBIMed, MedEvi, PICO Linguist, BabelMeSH, and PubCrawler aid in text crunching, retrieving evidence, multilanguage search, and staying updated with PubMed and GenBank.
– PubMed has various alternative interfaces accessible through platforms like Embase, Ovid, Dialog, and EBSCO, with over 500 licenses issued by NLM.