1
|
Mahmoudi S, Koch V, Santos DPD, Ackermann J, Grünewald LD, Weitkamp I, Yel I, Martin SS, Albrecht MH, Scholtz JE, Vogl TJ, Bernatz S. Imaging biomarkers to stratify lymph node metastases in abdominal CT - Is radiomics superior to dual-energy material decomposition? Eur J Radiol Open 2022; 10:100459. [PMID: 36561422 PMCID: PMC9763741 DOI: 10.1016/j.ejro.2022.100459] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/29/2022] [Revised: 11/16/2022] [Accepted: 11/30/2022] [Indexed: 12/13/2022] Open
Abstract
Purpose To assess the potential of radiomic features in comparison to dual-energy CT (DECT) material decomposition to objectively stratify abdominal lymph node metastases. Materials and methods In this retrospective study, we included 81 patients (m, 57; median age, 65 (interquartile range, 58.7-73.3) years) with either lymph node metastases (n = 36) or benign lymph nodes (n = 45) who underwent contrast-enhanced abdominal DECT between 06/2015-07/2019. All malignant lymph nodes were classified as unequivocal according to RECIST criteria and confirmed by histopathology, PET-CT or follow-up imaging. Three investigators segmented lymph nodes to extract DECT and radiomics features. Intra-class correlation analysis was applied to stratify a robust feature subset with further feature reduction by Pearson correlation analysis and LASSO. Independent training and testing datasets were applied on four different machine learning models. We calculated the performance metrics and permutation-based feature importance values to increase interpretability of the models. DeLong test was used to compare the top performing models. Results Distance matrices and t-SNE plots revealed clearer clusters using a combination of DECT and radiomic features compared to DECT features only. Feature reduction by LASSO excluded all DECT features of the combined feature cohort. The top performing radiomic features model (AUC = 1.000; F1 = 1.000; precision = 1.000; Random Forest) was significantly superior to the top performing DECT features model (AUC = 0.942; F1 = 0.762; precision = 0.800; Stochastic Gradient Boosting) (DeLong < 0.001). Conclusion Imaging biomarkers have the potential to stratify unequivocal lymph node metastases. Radiomics models were superior to DECT material decomposition and may serve as a support tool to facilitate stratification of abdominal lymph node metastases.
Collapse
Key Words
- ADB, AdaBoost
- AUC, Area under the curve
- Abdominal imaging
- CT, Computed tomography
- CTDI, Computed tomography dose index
- DECT, Dual-energy computed tomography
- DICOM, Digital Imaging and Communications in Medicine
- DLP, Dose-length product
- Dual-energy computed tomography
- GLCM, Gray Level Co-occurrence Matrix
- GLDM, Gray Level Dependence Matrix
- GLRLM, Gray Level Run Length Matrix
- GLSZM, Gray Level Size Zone Matrix
- HU, Hounsfield Units
- ICC, Intra-class correlation coefficient
- ID%, Normalized iodine uptake
- ID, Iodine density
- LR, Logistic Regression
- Lymph node metastasis
- Machine Learning
- NGTDM, Neighboring Gray Tone Difference Matrix
- Oncology
- PET, Positron emission tomography
- RF, Random Forest
- ROC, Receiver operating characteristics
- ROI, Region of interest
- Radiomics
- SGB, Stochastic Gradient Boosting
- VOI, Volume of interest
- mGy, Milligray
Collapse
Affiliation(s)
- Scherwin Mahmoudi
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany,Corresponding author.
| | - Vitali Koch
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Daniel Pinto Dos Santos
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany,University Hospital Cologne, Department of Diagnostic and Interventional Radiology, Kerpener Str. 62, 50937 Cologne, Germany
| | - Jörg Ackermann
- Department of Molecular Bioinformatics, Institute of Computer Science, Johann Wolfgang Goethe-University, Robert-Mayer-Str. 11-15, 60325 Frankfurt am Main, Germany
| | - Leon D. Grünewald
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Inga Weitkamp
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Ibrahim Yel
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Simon S. Martin
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Moritz H. Albrecht
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Jan-Erik Scholtz
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Thomas J. Vogl
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
| | - Simon Bernatz
- University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany,Dr. Senckenberg Institute for Pathology, University Hospital Frankfurt, Goethe University Frankfurt am Main, 60590 Frankfurt am Main, Germany
| |
Collapse
|