Diversity and Inclusion Within Datasets in Heart Failure
Bibliographic record
Abstract
BACKGROUND: Heart failure (HF) is a life-threatening disease affecting 64 million people worldwide. Artificial intelligence (AI) technologies are being developed for use in HF to support early diagnosis and stratification of treatment. The performance characteristics of AI technologies are influenced by whether the data used during the AI lifecycle reflects the populations for which the AI is used. OBJECTIVES: The aim of the study was to identify and characterize datasets used across the lifecycle of AI technologies for HF, focusing on data diversity and inclusivity. METHODS: MEDLINE and Embase were systematically searched from January 1, 2012, until August 30, 2022, to identify articles relating to the development of AI in HF. Articles were independently screened by 2 reviewers to identify datasets. Dataset documentation was analyzed with a focus on accessibility, geographical origin, relevant metadata reporting, and dataset composition. RESULTS: The 72 datasets identified represented 23 countries and over 2 million individuals. In total, 62 (86%) datasets reported "age," 61 (85%) reported sex or gender, 21 (29%) reported race and/or ethnicity, and 8 (11%) reported socioeconomic status. In the 21 datasets that reported race and/or ethnicity, 89% of individuals represented were reported within the "White" or "Caucasian" category. Only 20 (28%) datasets were fully accessible. CONCLUSIONS: Reporting of sex, gender, and socioeconomic status in HF datasets is inconsistent. There is a need to generate datasets that are transparently reported and accessible. Although collecting and reporting demographic attributes is complex and needs to be undertaken with appropriate safeguards, it is also an essential step toward building equitable AI-based health technologies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.437 | 0.744 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.005 | 0.006 |
| Bibliometrics | 0.049 | 0.055 |
| Science and technology studies | 0.004 | 0.011 |
| Scholarly communication | 0.013 | 0.027 |
| Open science | 0.006 | 0.019 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".