Healthsheet: Development of a Transparency Artifact for Health Datasets

Rostamzadeh, Negar; Mincu, Diana; Roy, Subhrajit; Smart, Andrew; Wilcox, Lauren; Pushkarna, Mahima; Schrouff, Jessica; Amironesei, Razvan; Moorosi, Nyalleng; Heller, Katherine

doi:10.1145/3531146.3533239

Cited by 35 publications

(17 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These cases, which are collated in the form of datasets, should adhere to emerging guidance as to what constitutes transparent and complete reporting. Derived from the seminal ‘Datasheets for Datasets’ effort, the recent ‘Healthsheet’ initiative has set forth healthcare-specific standards around the reporting of dataset transparency and diversity 35 , 36 . This initiative has also been endorsed by the ongoing STANDING Together project, which will further serve to provide guidance around minimising racial and ethical health inequalities in AI datasets dataset composition 37 .…”

Section: Discussionmentioning

confidence: 99%

The diagnostic and triage accuracy of digital and online symptom checker tools: a systematic review

et al. 2022

View full text Add to dashboard Cite

Digital and online symptom checkers are an increasingly adopted class of health technologies that enable patients to input their symptoms and biodata to produce a set of likely diagnoses and associated triage advice. However, concerns regarding the accuracy and safety of these symptom checkers have been raised. This systematic review evaluates the accuracy of symptom checkers in providing diagnoses and appropriate triage advice. MEDLINE and Web of Science were searched for studies that used either real or simulated patients to evaluate online or digital symptom checkers. The primary outcomes were the diagnostic and triage accuracy of the symptom checkers. The QUADAS-2 tool was used to assess study quality. Of the 177 studies retrieved, 10 studies met the inclusion criteria. Researchers evaluated the accuracy of symptom checkers using a variety of medical conditions, including ophthalmological conditions, inflammatory arthritides and HIV. A total of 50% of the studies recruited real patients, while the remainder used simulated cases. The diagnostic accuracy of the primary diagnosis was low across included studies (range: 19–37.9%) and varied between individual symptom checkers, despite consistent symptom data input. Triage accuracy (range: 48.8–90.1%) was typically higher than diagnostic accuracy. Overall, the diagnostic and triage accuracy of symptom checkers are variable and of low accuracy. Given the increasing push towards adopting this class of technologies across numerous health systems, this study demonstrates that reliance upon symptom checkers could pose significant patient safety hazards. Large-scale primary studies, based upon real-world data, are warranted to demonstrate the adequate performance of these technologies in a manner that is non-inferior to current best practices. Moreover, an urgent assessment of how these systems are regulated and implemented is required.

show abstract

Section: Discussionmentioning

confidence: 99%

The diagnostic and triage accuracy of digital and online symptom checker tools: a systematic review

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Problems arise when biases result in a lack of generalizability to a wider population, or when the importance of explanatory factors is not understood because particular variables are missing or underpowered (see also Challenge 8). As a start, introducing standardized processes for documenting missingness in datasets would help to ensure any conclusions appropriately acknowledge these biases 70,71 .…”

Section: Challenge 9: Ethical Implications Of Smmentioning

confidence: 99%

Learning from data with structured missingness

Mitra

McGough²,

Chakraborti

et al. 2023

Nat Mach Intell

View full text Add to dashboard Cite

“…The mere availability of this dataset during the early days of the COVID-19 pandemic caused the community to overindex without careful consideration of whether it was appropriate for their research question or not. Such cases could be avoided if datasets were accompanied by a transparency artefact (for example Healthsheet 20 ) along with the paper. The Healthsheet questionnaire, for example, contains specific questions on different aspects of a dataset such as the collection process, composition, distribution, cleaning and recommended use cases.…”

Section: Datasetsmentioning

confidence: 99%

“…• Release a transparency artefact by using standardized questionnaire templates (for example, Healthsheet 20 ) along with the paper.…”

Section: Recommendedmentioning

confidence: 99%