Categories | No. | Items | Weights | Scoref |
---|---|---|---|---|
Study design | #1 | Adherence to radiomics and/or machine learning-specific checklists or guidelines | 0.0368 | |
#2 | Eligibility criteria that describe a representative study population | 0.0735 | ||
#3 | High-quality reference standard with a clear definition | 0.0919 | ||
Imaging data | #4 | Multi-center | 0.0438 | |
#5 | Clinical translatability of the imaging data source for radiomics analysis | 0.0292 | ||
#6 | Imaging protocol with acquisition parameters | 0.0438 | ||
#7 | The interval between imaging used and reference standard | 0.0292 | ||
Segmentationa | #8 | Transparent description of segmentation methodology | 0.0337 | |
#9 | Formal evaluation of fully automated segmentationb | 0.0225 | ||
#10 | Test set segmentation masks produced by a single reader or automated tool | 0.0112 | ||
Image processing and feature extraction | #11 | Appropriate use of image preprocessing techniques with transparent description | 0.0622 | |
#12 | Use of standardized feature extraction softwarec | 0.0311 | ||
#13 | Transparent reporting of feature extraction parameters, otherwise providing a default configuration statement | 0.0415 | ||
Feature processing | #14 | Removal of non-robust featuresd | 0.0200 | |
#15 | Removal of redundant featuresd | 0.0200 | ||
#16 | Appropriateness of dimensionality compared to data sized | 0.0300 | ||
#17 | Robustness assessment of end-to-end deep learning pipelinese | 0.0200 | ||
Preparation for modeling | #18 | Proper data partitioning process | 0.0599 | |
#19 | Handling of confounding factors | 0.0300 | ||
Metrics and comparison | #20 | Use of appropriate performance evaluation metrics for task | 0.0352 | |
#21 | Consideration of uncertainty | 0.0234 | ||
#22 | Calibration assessment | 0.0176 | ||
#23 | Use of uni-parametric imaging or proof of its inferiority | 0.0117 | ||
#24 | Comparison with a non-radiomic approach or proof of added clinical value | 0.0293 | ||
#25 | Comparison with simple or classical statistical models | 0.0176 | ||
Testing | #26 | Internal testing | 0.0375 | |
#27 | External testing | 0.0749 | ||
Open science | #28 | Data availability | 0.0075 | |
#29 | Code availability | 0.0075 | ||
#30 | Model availability | 0.0075 | ||
Total METRICS score (should be given as percentage) | ||||
Quality categoryg |