Diagnostic Performance of 2D and 3D T2WI-Based Radiomics Features With Machine Learning Algorithms to Distinguish Solid Solitary Pulmonary Lesion
- PMID: 34868905
- PMCID: PMC8637439
- DOI: 10.3389/fonc.2021.683587
Diagnostic Performance of 2D and 3D T2WI-Based Radiomics Features With Machine Learning Algorithms to Distinguish Solid Solitary Pulmonary Lesion
Abstract
Objective: To evaluate the performance of 2D and 3D radiomics features with different machine learning approaches to classify SPLs based on magnetic resonance(MR) T2 weighted imaging (T2WI).
Material and methods: A total of 132 patients with pathologically confirmed SPLs were examined and randomly divided into training (n = 92) and test datasets (n = 40). A total of 1692 3D and 1231 2D radiomics features per patient were extracted. Both radiomics features and clinical data were evaluated. A total of 1260 classification models, comprising 3 normalization methods, 2 dimension reduction algorithms, 3 feature selection methods, and 10 classifiers with 7 different feature numbers (confined to 3-9), were compared. The ten-fold cross-validation on the training dataset was applied to choose the candidate final model. The area under the receiver operating characteristic curve (AUC), precision-recall plot, and Matthews Correlation Coefficient were used to evaluate the performance of machine learning approaches.
Results: The 3D features were significantly superior to 2D features, showing much more machine learning combinations with AUC greater than 0.7 in both validation and test groups (129 vs. 11). The feature selection method Analysis of Variance(ANOVA), Recursive Feature Elimination(RFE) and the classifier Logistic Regression(LR), Linear Discriminant Analysis(LDA), Support Vector Machine(SVM), Gaussian Process(GP) had relatively better performance. The best performance of 3D radiomics features in the test dataset (AUC = 0.824, AUC-PR = 0.927, MCC = 0.514) was higher than that of 2D features (AUC = 0.740, AUC-PR = 0.846, MCC = 0.404). The joint 3D and 2D features (AUC=0.813, AUC-PR = 0.926, MCC = 0.563) showed similar results as 3D features. Incorporating clinical features with 3D and 2D radiomics features slightly improved the AUC to 0.836 (AUC-PR = 0.918, MCC = 0.620) and 0.780 (AUC-PR = 0.900, MCC = 0.574), respectively.
Conclusions: After algorithm optimization, 2D feature-based radiomics models yield favorable results in differentiating malignant and benign SPLs, but 3D features are still preferred because of the availability of more machine learning algorithmic combinations with better performance. Feature selection methods ANOVA and RFE, and classifier LR, LDA, SVM and GP are more likely to demonstrate better diagnostic performance for 3D features in the current study.
Keywords: algorithms; area under the curve; lung neoplasms; machine learning; magnetic resonance imaging.
Copyright © 2021 Wan, Zhou, Xia, Hu, Wang, Peng, Zhang, Sun, Song, Yang and Li.
Conflict of interest statement
Authors TZ and JS were employed by Philips Healthcare. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Figures
![Figure 1](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g001.gif)
![Figure 2](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g002.gif)
![Figure 3](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g003.gif)
![Figure 4](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g004.gif)
![Figure 5](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g005.gif)
![Figure 6](https://www.ncbi.nlm.nih.gov/pmc/articles/instance/8637439/bin/fonc-11-683587-g006.gif)
Similar articles
-
Multimodality radiomics prediction of radiotherapy-induced the early proctitis and cystitis in rectal cancer patients: a machine learning study.Biomed Phys Eng Express. 2023 Dec 20;10(1). doi: 10.1088/2057-1976/ad0f3e. Biomed Phys Eng Express. 2023. PMID: 37995359
-
Biphasic majority voting-based comparative COVID-19 diagnosis using chest X-ray images.Expert Syst Appl. 2023 Apr 15;216:119430. doi: 10.1016/j.eswa.2022.119430. Epub 2022 Dec 21. Expert Syst Appl. 2023. PMID: 36570382 Free PMC article. Review.
-
Foundations of Machine Learning-Based Clinical Prediction Modeling: Part II-Generalization and Overfitting.Acta Neurochir Suppl. 2022;134:15-21. doi: 10.1007/978-3-030-85292-4_3. Acta Neurochir Suppl. 2022. PMID: 34862523 Review.
-
Classification of pulmonary lesion based on multiparametric MRI: utility of radiomics and comparison of machine learning methods.Eur Radiol. 2020 Aug;30(8):4595-4605. doi: 10.1007/s00330-020-06768-y. Epub 2020 Mar 28. Eur Radiol. 2020. PMID: 32222795
-
Radiomics analysis for the differentiation of autoimmune pancreatitis and pancreatic ductal adenocarcinoma in 18 F-FDG PET/CT.Med Phys. 2019 Oct;46(10):4520-4530. doi: 10.1002/mp.13733. Epub 2019 Aug 13. Med Phys. 2019. PMID: 31348535
Cited by
-
The value of machine learning based on CT radiomics in the preoperative identification of peripheral nerve invasion in colorectal cancer: a two-center study.Insights Imaging. 2024 Apr 5;15(1):101. doi: 10.1186/s13244-024-01664-1. Insights Imaging. 2024. PMID: 38578423 Free PMC article.
-
The effect of feature normalization methods in radiomics.Insights Imaging. 2024 Jan 7;15(1):2. doi: 10.1186/s13244-023-01575-7. Insights Imaging. 2024. PMID: 38185786 Free PMC article.
-
Artificial intelligence-driven radiomics study in cancer: the role of feature engineering and modeling.Mil Med Res. 2023 May 16;10(1):22. doi: 10.1186/s40779-023-00458-8. Mil Med Res. 2023. PMID: 37189155 Free PMC article. Review.
-
Collinearity and Dimensionality Reduction in Radiomics: Effect of Preprocessing Parameters in Hypertrophic Cardiomyopathy Magnetic Resonance T1 and T2 Mapping.Bioengineering (Basel). 2023 Jan 6;10(1):80. doi: 10.3390/bioengineering10010080. Bioengineering (Basel). 2023. PMID: 36671652 Free PMC article.
-
Biparametric magnetic resonance imaging-based radiomics features for prediction of lymphovascular invasion in rectal cancer.BMC Cancer. 2023 Jan 18;23(1):61. doi: 10.1186/s12885-023-10534-w. BMC Cancer. 2023. PMID: 36650498 Free PMC article.
References
-
- Meier-Schroers M, Homsi R, Gieseke J, Schild HH, Thomas D. Lung Cancer Screening With MRI: Evaluation of MRI for Lung Cancer Screening by Comparison of LDCT- and MRI-Derived Lung-RADS Categories in the First Two Screening Rounds. Eur Radiol (2019) 29:898–905. doi: 10.1007/s00330-018-5607-8 - DOI - PubMed
LinkOut - more resources
Full Text Sources
Research Materials