Large expert-curated database for benchmarking document similarity detection in biomedical literature search

Abstract

Document recommendation systems for locating relevant literature have mostly relied on methods developed a decade ago. This is largely due to the lack of a large offline gold-standard benchmark of relevant documents that cover a variety of research fields such that newly developed literature search techniques can be compared, improved and translated into practice. To overcome this bottleneck, we have established the RElevant LIterature SearcH consortium consisting of more than 1500 scientists from 84 countries, who have collectively annotated the relevance of over 180 000 PubMed-listed articles with regard to their respective seed (input) article/s. The majority of annotations were contributed by highly experienced, original authors of the seed articles. The collected data cover 76% of all unique PubMed Medical Subject Headings descriptors. No systematic biases were observed across different experience levels, research fields or time spent on annotations. More importantly, annotations of the same document pairs contributed by different scientists were highly concordant. We further show that the three representative baseline methods used to generate recommended articles for evaluation (Okapi Best Matching 25, Term Frequency–Inverse Document Frequency and PubMed Related Articles) had similar overall performances. Additionally, we found that these methods each tend to produce distinct collections of recommended articles, suggesting that a hybrid method may be required to completely capture all relevant articles. The established database server located at https://relishdb.ict.griffith.edu.au is freely available for the downloading of annotation data and the blind testing of new methods. We expect that this benchmark will be useful for stimulating the development of new powerful techniques for title and title/abstract-based search engines for relevant articles in biomedical research.

Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity

“…1. RELISH: An annotated dataset of biomedical abstract queries and candidate pools of size 60, labelled for similarity by experts who are often authors of the query abstracts (Brown et al, 2019).…”

“…Def 2. Aspect-level retrieval by sentences: Given query and candidate documents -Q and C, and a subset of sentences S Q ⊆ Q based on which to retrieve documents a system must output the ranking over C. Our definitions align with existing datasets and tasks of scientific document similarity: the well explored abstract-level similarity task (Brown et al, 2019;Cohan et al, 2020), and more recently introduced facet-based document similarity task (Mysore et al, 2021). §4 will describe evaluation on these datasets.…”

Psychophysiological Indicators of Postural Control. Contribution of the Russian Scientific School. Part I

“…The weighty contribution of the Russian scientific school was made at the very beginning of postural control studies for the objectification of the measurements of postural stability. According to the recommendations of the International Consensus on postural control measurements [ 27 ] established based on “Research Methods to Evaluate Standing Stability” developed in 1952 by Russian scientists [ 28 ], the efficiency of the balance maintenance system is assessed by measuring oscillations in the foot plantar center of pressure (COP) relative to the center of gravity (CG) [ 29 ]. These variations reflect the movements of body segments or joints, muscle activity, the movements associated with respiration [ 30 ], and the work of the cardiovascular system [ 31 ].…”

“…For reviewing the studies devoted to the psychophysiological mechanisms of postural control, literature data were searched by the following keywords: “postural control,” “sensorimotor integration,” sedentary lifestyle,” “gravitation,” “support,” “vestibular,” “proprioceptive” and “visual” “afferentation,” in combination with the keywords such as “cognitive functions,” “memory,” “attention,” “decision making,” “imagination,” “emotions,” “fine motor skills,” “dual tasks,” “anxiety,” “depression,” “stabilometry,” “Electroencephalography,” and “Electromyography” ( Table 1 ). Literature was searched in the Web of Science , PubMed , Scopus , and RSCI databases according to the recommendations of “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” ( PRISMA ) and using the methods described in RELISH ( RElevant LIterature SearcH ) [ 27 ]. The present review includes the results published in the articles with a Digital Object Identifier ( DOI ), completely corresponding to the keywords ( Table 1 ), except for those published only as abstracts.…”

Opportunities for Artificial Intelligence in Advancing Precision Medicine

“…ML can learn from almost any data type, even unstructured medical text, such as patient records, medical notes, prescriptions, audio interview transcripts, or pathology and radiology reports. Future day-to-day applications will embrace ML methods to organize a growing volume of scientific literature, facilitating access and extraction of meaningful knowledge content from it (24). In the clinic, ML can harness the potential of electronic health records to accurately predict medical events (25).…”