CLEAR item#51

“Practical implications. Summarize the practical implications of the results. Describe the key impact of the work on the field. Highlight the potential clinical value and role of the study. Discuss any issues that may hamper the successful translation of the study into real world clinical practice. Also, provide future expectations and possible next steps that others might build upon the current work.” [1] (from the article by Kocak et al.; licensed under CC BY 4.0)

Reporting examples for CLEAR item#51

Example#1. “[…] a model such as this implemented into the clinical routine pipeline may support inexperienced or moderately experienced radiologists or physicians in enhancing the quality of their decision-making regarding their diagnosis and consequently the further management or referral of these patients. More specifically, a model such as this may help with ‘ruling-in’ malignant lesions, particularly when the treating physician or the radiologist has limited experience. The patient could then be referred to a specialized center and a biopsy may be performed to secure the diagnosis.” [2] (from the article by von Schacky et al.; licensed under CC BY 4.0)

Example#2. “A positive result from the radiomic model could alert physicians to perform additional DEXA to confirm the presence of osteoporosis. In contrast, further testing and treatment may be unnecessary when the radiomic model shows a low probability of osteoporosis. This method may decrease medical costs and radiation exposure” [3] (from the article by Jiang et al.; licensed under CC BY 4.0)

Example#3. “First, our model was trained and validated using retrospective data collected from a single institution. A large-scale prospective and multicentre validation cohort collection is currently underway. Second, our radiomics analysis only predicted MGMT promoter methylation prediction from T1-CE, T2-FLAIR and ADC map images, which are the most common structural MR images. Additional scanning sequences such as dynamic susceptibility contrast, susceptibility-weighted imaging and diffusional kurtosis imaging will be included in future studies to further improve predictive performance. Third, the relationship between imaging features and critical molecular markers such as IDH and 1p19q should also be studied in future research. Finally, the manual segmentation method used to delineate ROIs in this study (tumour and oedema) was quite time consuming. Semi-automatic or deep learning-based automatic segmentation methods may enhance the objectivity of our method and promote the seamless integration of this technology into clinical application.” [4] (from the article by Wei et al.; licensed under CC BY 4.0)

Explanation and elaboration of CLEAR item#51

The ultimate goal of radiomics is to enhance clinical practice. Researchers must consider the clinical and practical aspects of their models early on. This approach not only indicates a clear pathway for further research but also ensures a smoother introduction and adoption of the model by practitioners. Therefore, it is valuable for every radiomics study to be accompanied by a detailed and critical discussion on the clinical value of the model, its role in routine clinical processes, and the limitations that need consideration or addressing in future investigations. Such discussions can assist in defining the necessary steps to translate the model into clinical practice. Example#1 and Example#2 above properly reveal the practical implications. On the other hand, despite being a limitations paragraph, Example#3 highlights the potential issues that may hamper the successful translation of the study into real-world clinical practice.

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. von Schacky CE, Wilhelm NJ, Schäfer VS, et al (2022) Development and evaluation of machine learning models based on X-ray radiomics for the classification and differentiation of malignant and benign bone tumors. Eur Radiol 32:6247–6257. https://doi.org/10.1007/s00330-022-08764-w
  3. Jiang Y-W, Xu X-J, Wang R, Chen C-M (2022) Radiomics analysis based on lumbar spine CT to detect osteoporosis. Eur Radiol 32:8019–8026. https://doi.org/10.1007/s00330-022-08805-4
  4. 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

Back Next