CLEAR item#40

“Uncertainty evaluation and measures (e.g., confidence intervals). Describe the uncertainty evaluation (e.g., robustness, sensitivity analysis, calibration analysis if applicable) and measures of uncertainty quantification (e.g., confidence intervals, standard deviation).” [1] (from the article by Kocak et al.; licensed under CC BY 4.0)

Reporting examples for CLEAR item#40

Example#1. “To assess predictive performance, we measured the area under the receiver operating curve (AUC). Additionally, we computed precision-recall AUC (PR AUC), accuracy, sensitivity, specificity, the F1 score, the negative predictive value (NPV), and the positive predictive value (PPV). Bootstrapping with 1000 samples was performed to compute 95% confidence intervals for all the evaluation metrics.” [2] (from the article by Bodalal et al.; licensed under CC BY 4.0)

Example#2. “Calibration for the prediction of tumour response to nCRT and FFDM was assessed via the Hosmer–Lemeshow goodness of fit test (HL test) and Greenwood Nam d’Agostino test (GND test), respectively.” [3] (from the article by Shahzadi et al.; licensed under CC BY 4.0)

Explanation and elaboration of CLEAR item#40

A thorough description of uncertainty evaluation is crucial to validate the reliability of results. Robustness, sensitivity analysis, and calibration analysis serve as essential components to assess the stability and generalizability of radiomic models [4]. Measures of uncertainty quantification, such as confidence intervals and standard deviation, play a crucial role in conveying the precision and reliability of the results. Incorporating these methods into radiomics research ensures a comprehensive evaluation of uncertainty, enhancing the robustness and applicability of the results.

References

  1. Kocak B, Baessler B, Bakas S, et al (2023) CheckList for EvaluAtion of Radiomics research (CLEAR): a step-by-step reporting guideline for authors and reviewers endorsed by ESR and EuSoMII. Insights Imaging 14:75. https://doi.org/10.1186/s13244-023-01415-8
  2. Bodalal Z, Bogveradze N, ter Beek LC, et al (2023) Radiomic signatures from T2W and DWI MRI are predictive of tumour hypoxia in colorectal liver metastases. Insights Imaging 14:133. https://doi.org/10.1186/s13244-023-01474-x
  3. Shahzadi I, Zwanenburg A, Lattermann A, et al (2022) Analysis of MRI and CT-based radiomics features for personalized treatment in locally advanced rectal cancer and external validation of published radiomics models. Sci Rep 12:10192. https://doi.org/10.1038/s41598-022-13967-8
  4. Park JE, Park SY, Kim HJ, Kim HS (2019) Reproducibility and Generalizability in Radiomics Modeling: Possible Strategies in Radiologic and Statistical Perspectives. Korean J Radiol 20:1124–1137. https://doi.org/10.3348/kjr.2018.0070

Back Next