Importance of sample size on the quality and utility of AI-based prediction models for healthcare
Bibliographic record
Abstract
Rigorous study design and analytical standards are required to generate reliable findings in healthcare from artificial intelligence (AI) research. One crucial but often overlooked aspect is the determination of appropriate sample sizes for studies developing AI-based prediction models for individual diagnosis or prognosis. Specifically, the number of participants and outcome events required in datasets for model training and evaluation remains inadequately addressed. Most AI studies do not provide a rationale for their chosen sample sizes and frequently rely on datasets that are inadequate for training or evaluating a clinical prediction model. Among the ten principles of Good Machine Learning Practice established by the US Food and Drug Administration, the UK Medicines and Healthcare products Regulatory Agency, and Health Canada, guidance on sample size is directly relevant to at least three principles. To reinforce this recommendation, we outline seven reasons why inadequate sample size negatively affects model training, evaluation, and performance. Using a range of examples, we illustrate these issues and discuss the potentially harmful consequences for patient care and clinical adoption. Additionally, we address challenges associated with increasing sample sizes in AI research and highlight existing approaches and software for calculating the minimum sample sizes required for model training and evaluation.
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 | 0.220 | 0.458 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.003 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.004 | 0.005 |
| Insufficient payload (model declined to judge) | 0.004 | 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".