1
|
Salimi M, Houshi S, Gholamrezanezhad A, Vadipour P, Seifi S. Radiomics-based machine learning in prediction of response to neoadjuvant chemotherapy in osteosarcoma: A systematic review and meta-analysis. Clin Imaging 2025; 123:110494. [PMID: 40349577 DOI: 10.1016/j.clinimag.2025.110494] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2025] [Revised: 05/03/2025] [Accepted: 05/07/2025] [Indexed: 05/14/2025]
Abstract
BACKGROUND AND AIMS Osteosarcoma (OS) is the most common primary bone malignancy, and neoadjuvant chemotherapy (NAC) improves survival rates. However, OS heterogeneity results in variable treatment responses, highlighting the need for reliable, non-invasive tools to predict NAC response. Radiomics-based machine learning (ML) offers potential for identifying imaging biomarkers to predict treatment outcomes. This systematic review and meta-analysis evaluated the accuracy and reliability of radiomics models for predicting NAC response in OS. METHODS A systematic search was conducted in PubMed, Embase, Scopus, and Web of Science up to November 2024. Studies using radiomics-based ML for NAC response prediction in OS were included. Pooled sensitivity, specificity, and AUC for training and validation cohorts were calculated using bivariate random-effects modeling, with clinical-combined models analyzed separately. Quality assessment was performed using the QUADAS-2 tool, radiomics quality score (RQS), and METRICS scores. RESULTS Sixteen studies were included, with 63 % using MRI and 37 % using CT. Twelve studies, comprising 1639 participants, were included in the meta-analysis. Pooled metrics for training cohorts showed an AUC of 0.93, sensitivity of 0.89, and specificity of 0.85. Validation cohorts achieved an AUC of 0.87, sensitivity of 0.81, and specificity of 0.82. Clinical-combined models outperformed radiomics-only models. The mean RQS score was 9.44 ± 3.41, and the mean METRICS score was 60.8 % ± 17.4 %. CONCLUSION Radiomics-based ML shows promise for predicting NAC response in OS, especially when combined with clinical indicators. However, limitations in external validation and methodological consistency must be addressed.
Collapse
Affiliation(s)
- Mohsen Salimi
- Research Center of Thoracic Oncology (RCTO), National Research Institute of Tuberculosis and Lung Diseases (NRITLD), Shahid Beheshti University of Medical Sciences, Tehran, Iran
| | - Shakiba Houshi
- Isfahan Neurosciences Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
| | - Ali Gholamrezanezhad
- Department of Radiology, Keck School of Medicine, University of Southern California (USC), Los Angeles, CA, USA
| | - Pouria Vadipour
- Western Michigan University Homer Stryker M.D. School of Medicine, Kalamazoo, MI, USA
| | - Sharareh Seifi
- Research Center of Thoracic Oncology (RCTO), National Research Institute of Tuberculosis and Lung Diseases (NRITLD), Shahid Beheshti University of Medical Sciences, Tehran, Iran.
| |
Collapse
|
2
|
Leng Y, Zhou J, Liu W, Luo F, Peng F, Gong L. A CT-based radiomics model for predicting progression-free survival in patients with epithelial ovarian cancer. BMC Cancer 2025; 25:899. [PMID: 40394512 PMCID: PMC12090431 DOI: 10.1186/s12885-025-14265-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/03/2025] [Accepted: 05/05/2025] [Indexed: 05/22/2025] Open
Abstract
PURPOSE This study aimed to develop and validate a CT-based radiomics nomogram for predicting the progression-free survival (PFS) of epithelial ovarian cancer (EOC). MATERIALS AND METHODS A total of 144 EOC patients were retrospectively enrolled from two hospitals and The Cancer Genome Atlas and The Cancer Imaging Archive, divided into a training set (n = 101) and a test set (n = 43) using a 7:3 ratio. Radiomic features were extracted from contrast enhanced CT images. The radiomics score (rad-score) was developed using the least absolute shrinkage and selection operator (LASSO) Cox regression. Clinical semantic features with P < 0.05 in multivariate Cox regression were combined with rad-score to develop radiomics nomogram. The predictive performance of the nomogram was assessed using the concordance index (C-index) and calibration curves. RESULTS Multivariate Cox regression analysis revealed that the International Federation of Obstetrics and Gynecology stage and residual tumor are significant predictors of PFS. Twelve radiomic features were selected by LASSO Cox regression. The combined model demonstrated superior predictive performance, with a C-index of 0.78 (95% CI: 0.689-0.889) in the training set, and 0.73 (95% CI: 0.572-0.886) in the test set. The combined model outperformed the clinical and radiomics models in predicting 1-, 3-, and 5-year PFS, with area under curves of 0.850 (95% CI: 0.722-0.943), 0.828 (95% CI: 0.722-0.901), and 0.845 (95% CI: 0.722-0.943), respectively. Calibration curves of the radiomic nomogram for prediction of 1-year, 3-year, 5-year PFS showed excellent calibrations in both training and test sets. CONCLUSION The combined model integrating rad-score and clinical semantic features effectively evaluates PFS in EOC patients. The radiomics nomogram provides a non-invasive, simple, and feasible method to predict PFS in EOC patients, which may facilitate clinical decision-making.
Collapse
Affiliation(s)
- Yinping Leng
- Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, Nanchang, Jiangxi Province, 330006, China
- Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, Nanchang, China
| | - Jingjing Zhou
- Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, Nanchang, Jiangxi Province, 330006, China
- Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, Nanchang, China
| | - Wenjie Liu
- Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, Nanchang, Jiangxi Province, 330006, China
- Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, Nanchang, China
| | - Fengyuan Luo
- Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, Nanchang, Jiangxi Province, 330006, China
- Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, Nanchang, China
| | - Fei Peng
- Department of Radiology, The First Affiliated Hospital, Hengyang Medical School, University of South China, Hengyang, China
| | - Lianggeng Gong
- Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University, Minde Road No. 1, Nanchang, Jiangxi Province, 330006, China.
- Jiangxi Provincial Key Laboratory of Intelligent Medical Imaging, Nanchang, China.
| |
Collapse
|
3
|
Kocak B, Ponsiglione A, Stanzione A, Ugga L, Klontzas ME, Cannella R, Cuocolo R. CLEAR guideline for radiomics: Early insights into current reporting practices endorsed by EuSoMII. Eur J Radiol 2024; 181:111788. [PMID: 39437630 DOI: 10.1016/j.ejrad.2024.111788] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/07/2024] [Revised: 10/07/2024] [Accepted: 10/12/2024] [Indexed: 10/25/2024]
Abstract
PURPOSE This study aims to evaluate current reporting practices in radiomics research, with a focus on CheckList for EvaluAtion of Radiomics research (CLEAR). METHODS We conducted a citation search using Google Scholar to collect original research articles on radiomics citing the CLEAR guideline up to June 17, 2024. We examined the adoption of the guideline, adherence scores per publication, item-wise adherence rates, and self-reporting practices. An expert panel from the European Society of Medical Imaging Informatics Radiomics Auditing Group conducted a detailed item-by-item confirmation analysis of the self-reported CLEAR checklists. RESULTS Out of 100 unique citations from 104 records, 48 original research papers on radiomics were included. The overall adoption rate in the literature was 2 %. Among the citing articles, 94 % (45/48) adopted CLEAR for reporting purposes, applying it to both hand-crafted radiomics (89 %) and deep learning (24 %). Self-reported checklists were included in 58 % (26/45) of these papers. Median study-wise adherence score for self-reported data was 91 % (interquartile range = 18 %). Mean confirmed adherence score was 66 % (standard deviation = 14 %). Difference between these scores was statistically significant, (mean = 21 %; standard deviation = 11 %), p < 0.001. Using an arbitrary 50 % adherence cut-off, the number of items with poor adherence increased from 3 to 15 after confirmation analysis, mostly comprised of open science-related items. In addition, several items were frequently misreported. CONCLUSION This study revealed significant discrepancies between self-reported and confirmed adherence to the CLEAR guideline in radiomics research, indicating a need for improved reporting accuracy and verification practices.
Collapse
Affiliation(s)
- Burak Kocak
- Department of Radiology, University of Health Sciences, Basaksehir Cam and Sakura City Hospital, Basaksehir, Istanbul, Türkiye.
| | - Andrea Ponsiglione
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy
| | - Arnaldo Stanzione
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy
| | - Lorenzo Ugga
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy
| | - Michail E Klontzas
- Artificial Intelligence and Translational Imaging (ATI) Laboratory, Department of Radiology, School of Medicine, University of Crete, Heraklion, Crete, Greece; Computational Biomedicine Lab, Institute of Computer Science, Foundation for Research and Technology (ICS-FORTH), Heraklion, Crete, Greece; Division of Radiology, Department of Clinical Science Intervention and Technology (CLINTEC), Karolinska Institute, Sweden
| | - Roberto Cannella
- Section of Radiology, Department of Biomedicine, Neuroscience and Advanced Diagnostics, University of Palermo, Palermo, Italy
| | - Renato Cuocolo
- Department of Medicine, Surgery and Dentistry, University of Salerno, Baronissi, Italy
| |
Collapse
|
4
|
Zhou S, Liu Q, Fu Y, Du L, Bao Q, Zhang Z, Xu Z, Yan F, Li M, Liu R, Qin L, Zhang W. CT-derived Radiomics Predicts the Efficacy of Tyrosine Kinase Inhibitors in Osteosarcoma Patients with Pulmonary Metastasis. Transl Oncol 2024; 45:101993. [PMID: 38743988 PMCID: PMC11109890 DOI: 10.1016/j.tranon.2024.101993] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/08/2024] [Revised: 04/02/2024] [Accepted: 05/07/2024] [Indexed: 05/16/2024] Open
Abstract
BACKGROUND To construct and validate the CT-based radiomics model for predicting the tyrosine kinase inhibitors (TKIs) effects in osteosarcoma (OS) patients with pulmonary metastasis. METHODS OS patients with pulmonary metastasis treated with TKIs were randomly separated into training and testing cohorts (2:1 ratio). Radiomic features were extracted from the baseline unenhanced chest CT images. The random survival forest (RSF) and Kaplan-Meier survival analyses were performed to construct and evaluate radiomics signatures (R-model-derived). The univariant and multivariant Cox regression analyses were conducted to establish clinical (C-model) and combined models (RC-model). The discrimination abilities, goodness of fit and clinical benefits of the three models were assessed and validated in both training and testing cohorts. RESULTS A total of 90 patients, 57 men and 33 women, with a mean age of 18 years and median progression-free survival (PFS) of 7.2 months, were enrolled. The R-model was developed with nine radiomic features and demonstrated significant predictive and prognostic values. In both training and testing cohorts, the time-dependent area under the receiver operating characteristic curves (AUC) of the R-model and RC-model exhibited obvious superiority over C-model. The calibration and decision curve analysis (DCA) curves indicated that the accuracy of the R-model was comparable to RC-model, which exhibited significantly better performance than C-model. CONCLUSIONS The R-model showed promising potential as a predictor for TKI responses in OS patients with pulmonary metastasis. It can potentially identify pulmonary metastatic OS patients most likely to benefit from TKIs treatment and help guide optimized clinical decisions.
Collapse
Affiliation(s)
- Shanshui Zhou
- Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China; Faculty of Medical Imaging Technology, College of Health Science and Technology, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Qi Liu
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Yucheng Fu
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Lianjun Du
- Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Qiyuan Bao
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Zhusheng Zhang
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Zhihan Xu
- Siemens Healthineers CT Collaboration, Shanghai, PR China
| | - Fuhua Yan
- Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China; Faculty of Medical Imaging Technology, College of Health Science and Technology, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Meng Li
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Ruixuan Liu
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Le Qin
- Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China
| | - Weibin Zhang
- Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, PR China.
| |
Collapse
|