CLEAR item#42

“Comparison with non-radiomic and combined methods. Indicate whether comparisons with nonradiomic approaches (e.g., clinical parameters, laboratory parameters, traditional radiological evaluations) are performed. Non-radiomic approaches can be combined with radiomic data as well (e.g., clinical-radiomic evaluation). Explain how the clinical utility is assessed, such as with decision curve analysis.” [1] (from the article by Kocak et al.; licensed under CC BY 4.0)

Reporting examples for CLEAR item#42

Example#1. “To achieve a holistic information-gathered network, we generated a comprehensive model including the fusion radiomics signature, the selected clinical factor (oedema degree), and two ADC values (the tumour and oedema areas). […] We chose decision curve analysis (DCA) to estimate the clinical usefulness of the developed fusion radiomics signature and used the Delong test to explore whether the fusion radiomics signature performed better than the traditional clinical model and ADC parameter.” [2] (from the article by Wei et al.; licensed under CC BY 4.0)

Example#2. “Moreover, the two classifiers were built, including patient’s age and sex along with the selected radiomic features, to verify if their contribution improves the classification performance.” [3] (from the article by Doniselli et al.; licensed under CC BY 4.0)

Explanation and elaboration of CLEAR item#42

A comparison of the performance of radiomics studies and studies without the use of artificial intelligence is required to confirm or deny an advantage in the use of such new approaches. According to a recent meta-research, about half of the radiomic studies lack such an important comparison [4]. The use of non-radiomic variables in classification algorithms can improve performance. Such variables may be easy to implement, such as demographic variables like gender and age. Decision curve analysis can be used to estimate the net clinical benefit of the predictive/diagnostic model but is rarely used in the medical literature [5]. It should also be noted that this tool is often misused in the literature, although it can still provide useful insights [6].

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. Wei J, Yang G, Hao X, et al (2019) A multi-sequence and habitat-based MRI radiomics signature for preoperative prediction of MGMT promoter methylation in astrocytomas with prognostic implication. Eur Radiol 29:877–888. https://doi.org/10.1007/s00330-018-5575-z
  3. Doniselli FM, Pascuzzo R, Agrò M, et al (2024) Development of A Radiomic Model for MGMT Promoter Methylation Detection in Glioblastoma Using Conventional MRI. Int J Mol Sci 25:138. https://doi.org/10.3390/ijms25010138
  4. Kocak B, Bulut E, Bayrak ON, et al (2023) NEgatiVE results in Radiomics research (NEVER): A meta-research study of publication bias in leading radiology journals. Eur J Radiol 163:110830. https://doi.org/10.1016/j.ejrad.2023.110830
  5. Vickers AJ, van Calster B, Steyerberg EW (2019) A simple, step-by-step guide to interpreting decision curve analysis. Diagn Progn Res 3:18. https://doi.org/10.1186/s41512-019-0064-7
  6. Capogrosso P, Vickers AJ (2019) A Systematic Review of the Literature Demonstrates Some Errors in the Use of Decision Curve Analysis but Generally Correct Interpretation of Findings. Med Decis Making 39:493–498. https://doi.org/10.1177/0272989X19832881

Back Next