Table 2.
Overview of major reporting guidelines and bias assessment tools for diagnostic and prognostic studies.
| Guideline/Tool | Full Name | Year Published | Articles Targeted | Purpose | Specific to ML? |
|---|---|---|---|---|---|
| QUADAS-2 4 | Quality Assessment of Diagnostic Accuracy Studies | 2011 (original QUADAS 4: 2003) | Diagnostic accuracy studies | Evaluates study risk of bias and applicability | No; QUADAS-AI 4 is in development |
| TRIPOD 6 | Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis | 2015 | Studies developing, validating, or updating a diagnostic or prognostic prediction model | Provides a set of recommendations for study reporting | No; TRIPOD-AI 6 is in development |
| RQS 5 | Radiomics quality score | 2017 | Radiomic studies | Assesses study quality (emulating TRIPOD 6) | No |
| PROBAST 3 | Prediction model Risk Of Bias ASsessment Tool | 2019 | Studies developing, validating, or updating a diagnostic or prognostic prediction model | Evaluates study risk of bias and applicability | No; PROBAST-AI 3 is in development |
| CLAIM 2 | Checklist for AI 1 in Medical Imaging | 2020 | AI 1 studies in medical imaging | Guides authors in presenting (and aids reviewers in evaluating) their research | Yes |
1 AI = artificial intelligence, 2 CLAIM = Checklist for AI in Medical Imaging, 3 PROBAST = Prediction model Risk Of Bias ASsessment Tool, 4 QUADAS-2 = Quality Assessment of Diagnostic Accuracy Studies, 5 RQS = radiomics quality score, and 6 TRIPOD = Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis.