User needs analysis and usability assessment of DataMed – a biomedical data discovery index

Our manual review shows that the ingestion pipeline could achieve an accuracy of 90% and core elements of DATS had varied frequency across repositories. On a manually curated benchmark dataset, the DataMed search engine achieved an inferred average precision of 0.2033 and a precision at 10 (P@10, the number of relevant results in the top 10 search results) of 0.6022, by implementing advanced natural language processing and terminology services. Currently, we have made the DataMed system publically available as an open source package for the biomedical community.

show abstract

“…To better understand users’ needs, usability studies were also conducted, providing guidance for the iterative development of DataMed. 38 …”

Section: Resultsmentioning

confidence: 99%

DataMed – an open source discovery index for finding biomedical datasets

Chen

Gururaj

Özyurt

et al. 2018

Journal of the American Medical Informatics Association

Self Cite

View full text Add to dashboard Cite

show abstract

“…The use of metadata (structured data about the data) to assist users with discovering and accessing open health data is not well studied or understood. Interestingly, Dixit et al (2018) found the most significant issue with usability of a dataset was incomplete, inconsistent, and poor-quality metadata [ 2 ]. The impact of open health data is impeded by poor metadata practices when such practices make the datasets difficult to discover and access for the various interested audiences.…”

Section: Introductionmentioning

confidence: 99%

“…Most of the research on open health data repositories has occurred in the last five years, which indicates its importance as an emerging field of study. Evaluations of metadata in open health data repositories focussed on adherence to the Dublin Core (DC) metadata standard [ 7 , 8 ] and the Open Archive Initiative-Protocol for Metadata Harvesting (OAI-PMH) interoperability standard [ 9 ], and metadata for datasets in the repositories [ 2 , 10 - 12 ]. The DC metadata standard is a simple and effective set of elements to describe various networked resources [ 13 ].…”

Section: Introductionmentioning

confidence: 99%

Challenges with organization, discoverability and access in Canadian open health data repositories

Thornton

Shiri²

2021

J Can Health Libr Assoc

View full text Add to dashboard Cite

Introduction: Open health data provides healthcare professionals, biomedical researchers and the general public with access to health data which has the potential to improve healthcare delivery and policy. The challenge for data providers is to create and implement appropriate metadata, or structured data about the data, to ensure that data are easy to discover, access and re-use. The goal of this study is to identify, evaluate and compare Canadian open health data repositories for their searching, browsing and navigation functionalities, the richness of their metadata description practices, and their metadata-based filtering mechanisms.Methods: Metadata-based search and browsing was evaluated in addition to the number and nature of metadata elements. Canadian open health data repositories across national, provincial and institutional levels were evaluated. Data collected using verbatim text recording was evaluated using an analytical framework based on the 2019 Dataverse North Metadata Best Practices guide and 2019 Data Citation Implementation Project roadmap.Results: All six repositories required filtering to access “open health data”. All six repositories included subject facets for filtering, and title and description on the Results List. Inconsistencies suggest that improvements should address advanced search, health-specific search terms, records for all repositories and links to related publications.Discussion: Consistent use of title and description suggests that an interoperable interface is possible. Records indicate the need for explicit, easy to find mechanisms to access metadata in repositories. The analytical framework represents first draft guidelines for metadata creation and implementation to improve organization, discoverability and access to Canadian open health data.

show abstract

“…These practices have raised questions regarding the trustworthiness of the results obtained from the basic analysis of such mixed datasets. As a result, a focus on data validation, quality, and deeper evaluation has more recently emerged with the extreme growth in the volume of biological and biomedical data [8,20,27]. As bioinformatics transitions towards a second generation of advanced studies, there is need to focus on data quality issues as well as sophisticated data analytics techniques.…”

Section: Introduction 1background and Motivationmentioning

confidence: 99%