CLEAR item#2

“Structured summary with relevant information. Provide a structured summary of the purpose, methods, results, and conclusions, presenting only the most important aspects directly related to the purpose of the study. The abstract should be understandable on its own, without reading the main text. Considering the submission guidelines of the journals, it is recommended to specify the following items: the baseline characteristics (e.g., number of patients, scans, images, classes), data source (e.g., public, institutional), study nature (e.g., prospective, retrospective), segmentation technique (e.g., automated, semi-automated, or manual), feature extraction technique (e.g., hand-crafted, engineered, deep), dimensionality reduction techniques (e.g., feature selection, reproducibility analysis, multi-collinearity), modeling details (e.g., algorithms/models), validation technique (e.g., cross-validation), unseen testing (internal hold-out, external testing), model performance metrics (e.g., the area under the curve) with uncertainty measures (e.g., confidence intervals), number of the final set of features, traditional statistical methods with p-values, and open science status (e.g., public availability of data, code, and/or model).” [1] (from the article by Kocak et al.; licensed under CC BY 4.0)

Reporting examples for CLEAR item#2

Example#1. “Background. Solid components of part-solid nodules (PSNs) at CT are reflective of invasive adenocarcinoma, but studies describing radiomic features of PSNs and the perinodular region are lacking. Purpose. To develop and to validate radiomic signatures diagnosing invasive lung adenocarcinoma in PSNs compared with the Brock, clinical-semantic features, and volumetric models. Materials and Methods. This retrospective multicenter study (https://ClinicalTrials.gov, NCT03872362) included 291 patients (median age, 60 years; interquartile range, 55-65 years; 191 women) from January 2013 to October 2017 with 297 PSN lung adenocarcinomas split into training (n = 229) and test (n = 68) data sets. Radiomic features were extracted from the different regions (gross tumor volume [GTV], solid, ground-glass, and perinodular). Random-forest models were trained using clinical-semantic, volumetric, and radiomic features, and an online nodule calculator was used to compute the Brock model. Performances of models were evaluated using standard metrics such as area under the curve (AUC), accuracy, and calibration. The integrated discrimination improvement was applied to assess model performance changes after the addition of perinodular features. Results. The radiomics model based on ground-glass and solid features yielded an AUC of 0.98 (95% confidence interval [CI]: 0.96, 1.00) on the test data set, which was significantly higher than the Brock (AUC, 0.83 [95% CI: 0.72, 0.94]; P = .007), clinical-semantic (AUC, 0.90 [95% CI: 0.83, 0.98]; P = .03), volumetric GTV (AUC, 0.87 [95% CI: 0.78, 0.96]; P = .008), and radiomics GTV (AUC, 0.88 [95% CI: 0.80, 0.96]; P = .01) models. It also achieved the best accuracy (93% [95% CI: 84%, 98%]). Both this model and the model with added perinodular features showed good calibration, whereas adding perinodular features did not improve the performance (integrated discrimination improvement, -0.02; P = .56). Conclusion. Separating ground-glass and solid CT radiomic features of part-solid nodules was useful in diagnosing the invasiveness of lung adenocarcinoma, yielding a better predictive performance than the Brock, clinical-semantic, volumetric, and radiomics gross tumor volume models.” [2] (from the article by Wu et al.; licensed under CC BY 4.0)

Example#2. “Objectives: To build a machine learning (ML) model to detect extraprostatic extension (EPE) of prostate cancer (PCa), based on radiomics features extracted from prostate MRI index lesions. Methods: Consecutive MRI exams of patients undergoing radical prostatectomy for PCa were retrospectively collected from three institutions. Axial T2-weighted and apparent diffusion coefficient map images were annotated to obtain index lesion volumes of interest for radiomics feature extraction. Data from one institution was used for training, feature selection (using reproducibility, variance and pairwise correlation analyses, and a correlation-based subset evaluator), and tuning of a support vector machine (SVM) algorithm, with stratified 10-fold cross-validation. The model was tested on the two remaining institutions’ data and compared with a baseline reference and expert radiologist assessment of EPE. Results: In total, 193 patients were included. From an initial dataset of 2436 features, 2287 were excluded due to either poor stability, low variance, or high collinearity. Among the remaining, 14 features were used to train the ML model, which reached an overall accuracy of 83% in the training set. In the two external test sets, the SVM achieved an accuracy of 79% and 74% respectively, not statistically different from that of the radiologist (81-83%, p = 0.39-1) and outperforming the baseline reference (p = 0.001-0.02). Conclusions: A ML model solely based on radiomics features demonstrated high accuracy for EPE detection and good generalizability in a multicenter setting. Paired to qualitative EPE assessment, this approach could aid radiologists in this challenging task.” [3] (from the article by Cuocolo et al.; licensed under CC BY 4.0)

Explanation and elaboration of CLEAR item#2

The abstract is one of the most crucial parts of the paper. This short text must include the main and relevant information to make the paper quickly and effectively comparable to other publications without reading the main text [4]. Remember that the abstract may be the only content available to some readers or one of the first parts of the paper evaluated by a review for publication, so it is important to clearly state the study endpoints and analytical methods. A brief abstract that provides key project information should help readers decide whether to read the full text. This is especially important for radiomics publications, where the dataset and analysis indicate the paper’s quality and evidence. Thus, it is necessary to provide the main key points about the patient sample (e.g., as in Example#1) and the radiomics analysis steps (e.g., image preprocessing, segmentation, feature extraction, and analysis; as in Example#2) that directly affect conclusions [5]. Example#1 also summarizes the key details about the uncertainty of performance metrics and statistical comparisons in the results section.

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. Wu G, Woodruff HC, Shen J, et al (2020) Diagnosis of Invasive Lung Adenocarcinoma Based on Chest CT Radiomic Features of Part-Solid Pulmonary Nodules: A Multicenter Study. Radiology 297:451–458. https://doi.org/10.1148/radiol.2020192431
  3. Cuocolo R, Stanzione A, Faletti R, et al (2021) MRI index lesion radiomics and machine learning for detection of extraprostatic extension of disease: a multicenter study. Eur Radiol 31:7575–7583. https://doi.org/10.1007/s00330-021-07856-3
  4. Cals JWL, Kotz D (2013) Effective writing and publishing scientific papers, part II: title and abstract. J Clin Epidemiol 66:585. https://doi.org/10.1016/j.jclinepi.2013.01.005
  5. Beller EM, Glasziou PP, Altman DG, et al (2013) PRISMA for Abstracts: Reporting Systematic Reviews in Journal and Conference Abstracts. PLoS Med 10:e1001419. https://doi.org/10.1371/journal.pmed.1001419

Back Next