Prediction of High-Risk Neuroblastoma Among Neuroblastic Tumors Using Radiomics Features Derived from Magnetic Resonance Imaging: A Pilot Study
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Purpose: This study aimed to predict high-risk neuroblastoma among neuroblastic tumors using radiomics features extracted from MRI. Materials and Methods: Pediatric patients (age≤18 years) diagnosed with neuroblastic tumors who had pre-treatment MR images available were enrolled from institution A from January 2010 to November 2019 (training set) and institution B from January 2016 to January 2022 (test set). Segmentation was performed with regions of interest manually drawn along tumor margins on the slice with the widest tumor area by two radiologists. First-order and texture features were extracted and intraclass correlation coefficients (ICCs) were calculated. Multivariate logistic regression (MLR) and random forest (RF) models from 10-fold cross-validation were built using these features. The trained MLR and RF models were tested in an external test set. Results: Thirty-two patients (M:F=23:9, 26.0±26.7 months) were in the training set and 14 patients (M:F=10:4, 33.4±20.4 months) were in the test set with radiomics features (n=930) being extracted. For 10 of the most relevant features selected, intra- and interobserver variability was moderate to excellent (ICCs 0.633–0.911, 0.695–0.985, respectively). The area under the receiver operating characteristic curve (AUC) was 0.94 (sensitivity 67%, specificity 91%, and accuracy 84%) for the MLR model and the average AUC was 0.83 (sensitivity 44%, specificity 87%, and accuracy 75%) for the RF model from 10-fold cross-validation. In the test set, AUCs of the MLR and RF models were 0.94 and 0.91, respectively. Conclusion: An MRI-based radiomics model can help predict high-risk neuroblastoma among neuroblastic tumors.
Abstract
Purpose: This study aimed to predict high-risk neuroblastoma among neuroblastic tumors using radiomics features extracted from MRI. Materials and Methods: Pediatric patients (age≤18 years) diagnosed with neuroblastic tumors who had pre-treatment MR images available were enrolled from institution A from January 2010 to November 2019 (training set) and institution B from January 2016 to January 2022 (test set). Segmentation was performed with regions of interest manually drawn along tumor margins on the slice with the widest tumor area by two radiologists. First-order and texture features were extracted and intraclass correlation coefficients (ICCs) were calculated. Multivariate logistic regression (MLR) and random forest (RF) models from 10-fold cross-validation were built using these features. The trained MLR and RF models were tested in an external test set. Results: Thirty-two patients (M:F=23:9, 26.0±26.7 months) were in the training set and 14 patients (M:F=10:4, 33.4±20.4 months) were in the test set with radiomics features (n=930) being extracted. For 10 of the most relevant features selected, intra- and interobserver variability was moderate to excellent (ICCs 0.633–0.911, 0.695–0.985, respectively). The area under the receiver operating characteristic curve (AUC) was 0.94 (sensitivity 67%, specificity 91%, and accuracy 84%) for the MLR model and the average AUC was 0.83 (sensitivity 44%, specificity 87%, and accuracy 75%) for the RF model from 10-fold cross-validation. In the test set, AUCs of the MLR and RF models were 0.94 and 0.91, respectively. Conclusion: An MRI-based radiomics model can help predict high-risk neuroblastoma among neuroblastic tumors.
참고문헌 (24)
- Lonergan GJ, "Neuroblastoma, ganglioneuroblastoma, and ganglioneuroma : radiologic-pathologic correlation", Radiographics, 22, pp.911~934 (2002)
- He WG, "Clinical and biological features of neuroblastic tumors : a comparison of neuroblastoma and ganglioneuroblastoma", Oncotarget, 8, pp.37730~37739 (2017)
- Monclair T, "The International Neuroblastoma Risk Group(INRG)staging system : an INRG task force report", J Clin Oncol, 27, pp.298~303 (2009)
- Pinto NR, "Advances in risk classification and treatment strategies for neuroblastoma", J Clin Oncol, 33, pp.3008~3017 (2015)
- Shur JD, "Radiomics in oncology : a practical guide", Radiographics, 41, pp.1717~1732 (2021)
- van Timmeren JE, "Radiomics in medical imaging-"how-to"guide and critical reflection", Insights Imaging, 11, pp.91~ (2020)
- Meng Y, "Application of radiomics for personalized treatment of cancer patients", Cancer Manag Res, 11, pp.10851~10858 (2019)
- Dong Y, "Preoperative prediction of sentinel lymph node metastasis in breast cancer based on radiomics of T2-weighted fat-suppression and diffusionweighted MRI", Eur Radiol, 28, pp.582~591 (2018)
- Zhang P, "T2-weighted image-based radiomics signature for discriminating between seminomas and nonseminoma", Front Oncol, 9, pp.1330~ (2019)
- Liu X, "Two-dimensional and three-dimensional T2 weighted imaging-based radiomic signatures for the preoperative discrimination of ovarian borderline tumors and malignant tumors", J Ovarian Res, 15, pp.22~ (2022)
- Brisse HJ, "Radiogenomics of neuroblastomas : relationships between imaging phenotypes, tumor genomic profile and survival", PLoS One, 12, pp.e0185190~ (2017)
- Chen X, "CTbased radiomics signature with machine learning predicts MYCN amplification in pediatric abdominal neuroblastoma", Front Oncol, 11, pp.687884~ (2021)
- Collewet G, "Influence of MRI acquisition protocols and image intensity normalization methods on texture classification", Magn Reson Imaging, 22, pp.81~91 (2004)
- Carré A, "Standardization of brain MR images across machines and protocols : bridging the gap for MRI-based radiomics", Sci Rep, 10, pp.12340~ (2020)
- Park SH, "Robustness of magnetic resonance radiomic features to pixel size resampling and interpolation in patients with cervical cancer", Cancer Imaging, 21, pp.19~ (2021)
- Wu H, "Radiogenomics of neuroblastoma in pediatric patients : CT-based radiomics signature in predicting MYCN amplification", Eur Radiol, 31, pp.3080~3089 (2021)
- Di Giannatale A, "Radiogenomics prediction for MYCN amplification in neuroblastoma : a hypothesis generating study", Pediatr Blood Cancer, 68, pp.e29110~ (2021)
- Feng L, "Prediction for mitosis-karyorrhexis index status of pediatric neuroblastoma via machine learning based 18F-FDG PET", CT radiomics, 2022, pp.12, 262~
- Saha A, "Breast cancer MRI radiomics : an overview of algorithmic features and impact of interreader variability in annotating tumors", Med Phys, 45, pp.3076~3085 (2018)
- Hectors SJ, "MRI radiomics features predict immuno-oncological characteristics of hepatocellular carcinoma", Eur Radiol, 30, pp.3759~3769 (2020)
- Sun Y, "Multiparametric MRI and radiomics in prostate cancer : a review", Australas Phys Eng Sci Med, 42, pp.3~25 (2019)
- Scalco E, "T2w-MRI signal normalization affects radiomics features reproducibility", Med Phys, 47, pp.1680~1691 (2020)
- George RE, "Tumor histology during induction therapy in patients with high-risk neuroblastoma", Pediatr Blood Cancer, 59, pp.506~510 (2012)
- Kembhavi SA, "Imaging in neuroblastoma : an update", Indian J Radiol Imaging, 25, pp.129~136 (2015)