RESUMEN
BACKGROUND AND PURPOSE: Pediatric supratentorial tumors such as embryonal tumors, high-grade gliomas, and ependymomas are difficult to distinguish by histopathology and imaging because of overlapping features. We applied machine learning to uncover MR imaging-based radiomics phenotypes that can differentiate these tumor types. MATERIALS AND METHODS: Our retrospective cohort of 231 patients from 7 participating institutions had 50 embryonal tumors, 127 high-grade gliomas, and 54 ependymomas. For each tumor volume, we extracted 900 Image Biomarker Standardization Initiative-based PyRadiomics features from T2-weighted and gadolinium-enhanced T1-weighted images. A reduced feature set was obtained by sparse regression analysis and was used as input for 6 candidate classifier models. Training and test sets were randomly allocated from the total cohort in a 75:25 ratio. RESULTS: The final classifier model for embryonal tumor-versus-high-grade gliomas identified 23 features with an area under the curve of 0.98; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.85, 0.91, 0.79, 0.94, and 0.89, respectively. The classifier for embryonal tumor-versus-ependymomas identified 4 features with an area under the curve of 0.82; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.93, 0.69, 0.76, 0.90, and 0.81, respectively. The classifier for high-grade gliomas-versus-ependymomas identified 35 features with an area under the curve of 0.96; the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.82, 0.94, 0.82, 0.94, and 0.91, respectively. CONCLUSIONS: In this multi-institutional study, we identified distinct radiomic phenotypes that distinguish pediatric supratentorial tumors, high-grade gliomas, and ependymomas with high accuracy. Incorporation of this technique in diagnostic algorithms can improve diagnosis, risk stratification, and treatment planning.
Asunto(s)
Neoplasias Encefálicas , Ependimoma , Glioma , Neoplasias de Células Germinales y Embrionarias , Tumores Neuroectodérmicos Primitivos , Neoplasias Supratentoriales , Neoplasias Encefálicas/genética , Niño , Ependimoma/diagnóstico por imagen , Glioma/genética , Humanos , Imagen por Resonancia Magnética/métodos , Neoplasias de Células Germinales y Embrionarias/diagnóstico por imagen , Estudios Retrospectivos , Neoplasias Supratentoriales/diagnóstico por imagenRESUMEN
BACKGROUND AND PURPOSE: Atypical teratoid/rhabdoid tumors and medulloblastomas have similar imaging and histologic features but distinctly different outcomes. We hypothesized that they could be distinguished by MR imaging-based radiomic phenotypes. MATERIALS AND METHODS: We retrospectively assembled T2-weighted and gadolinium-enhanced T1-weighted images of 48 posterior fossa atypical teratoid/rhabdoid tumors and 96 match-paired medulloblastomas from 7 institutions. Using a holdout test set, we measured the performance of 6 candidate classifier models using 6 imaging features derived by sparse regression of 900 T2WI and 900 T1WI Imaging Biomarker Standardization Initiative-based radiomics features. RESULTS: From the originally extracted 1800 total Imaging Biomarker Standardization Initiative-based features, sparse regression consistently reduced the feature set to 1 from T1WI and 5 from T2WI. Among classifier models, logistic regression performed with the highest AUC of 0.86, with sensitivity, specificity, accuracy, and F1 scores of 0.80, 0.82, 0.81, and 0.85, respectively. The top 3 important Imaging Biomarker Standardization Initiative features, by decreasing order of relative contribution, included voxel intensity at the 90th percentile, inverse difference moment normalized, and kurtosis-all from T2WI. CONCLUSIONS: Six quantitative signatures of image intensity, texture, and morphology distinguish atypical teratoid/rhabdoid tumors from medulloblastomas with high prediction performance across different machine learning strategies. Use of this technique for preoperative diagnosis of atypical teratoid/rhabdoid tumors could significantly inform therapeutic strategies and patient care discussions.
Asunto(s)
Neoplasias Cerebelosas , Meduloblastoma , Tumor Rabdoide , Humanos , Imagen por Resonancia Magnética , Meduloblastoma/diagnóstico por imagen , Fenotipo , Estudios Retrospectivos , Tumor Rabdoide/diagnóstico por imagenRESUMEN
BACKGROUND AND PURPOSE: Posterior fossa tumors are the most common pediatric brain tumors. MR imaging is key to tumor detection, diagnosis, and therapy guidance. We sought to develop an MR imaging-based deep learning model for posterior fossa tumor detection and tumor pathology classification. MATERIALS AND METHODS: The study cohort comprised 617 children (median age, 92 months; 56% males) from 5 pediatric institutions with posterior fossa tumors: diffuse midline glioma of the pons (n = 122), medulloblastoma (n = 272), pilocytic astrocytoma (n = 135), and ependymoma (n = 88). There were 199 controls. Tumor histology served as ground truth except for diffuse midline glioma of the pons, which was primarily diagnosed by MR imaging. A modified ResNeXt-50-32x4d architecture served as the backbone for a multitask classifier model, using T2-weighted MRIs as input to detect the presence of tumor and predict tumor class. Deep learning model performance was compared against that of 4 radiologists. RESULTS: Model tumor detection accuracy exceeded an AUROC of 0.99 and was similar to that of 4 radiologists. Model tumor classification accuracy was 92% with an F1 score of 0.80. The model was most accurate at predicting diffuse midline glioma of the pons, followed by pilocytic astrocytoma and medulloblastoma. Ependymoma prediction was the least accurate. Tumor type classification accuracy and F1 score were higher than those of 2 of the 4 radiologists. CONCLUSIONS: We present a multi-institutional deep learning model for pediatric posterior fossa tumor detection and classification with the potential to augment and improve the accuracy of radiologic diagnosis.
Asunto(s)
Aprendizaje Profundo , Interpretación de Imagen Asistida por Computador/métodos , Neoplasias Infratentoriales/clasificación , Neoplasias Infratentoriales/diagnóstico por imagen , Adolescente , Adulto , Niño , Preescolar , Femenino , Humanos , Lactante , Neoplasias Infratentoriales/patología , Imagen por Resonancia Magnética/métodos , Masculino , Adulto JovenRESUMEN
BACKGROUND AND PURPOSE: Distinct molecular subgroups of pediatric medulloblastoma confer important differences in prognosis and therapy. Currently, tissue sampling is the only method to obtain information for classification. Our goal was to develop and validate radiomic and machine learning approaches for predicting molecular subgroups of pediatric medulloblastoma. MATERIALS AND METHODS: In this multi-institutional retrospective study, we evaluated MR imaging datasets of 109 pediatric patients with medulloblastoma from 3 children's hospitals from January 2001 to January 2014. A computational framework was developed to extract MR imaging-based radiomic features from tumor segmentations, and we tested 2 predictive models: a double 10-fold cross-validation using a combined dataset consisting of all 3 patient cohorts and a 3-dataset cross-validation, in which training was performed on 2 cohorts and testing was performed on the third independent cohort. We used the Wilcoxon rank sum test for feature selection with assessment of area under the receiver operating characteristic curve to evaluate model performance. RESULTS: Of 590 MR imaging-derived radiomic features, including intensity-based histograms, tumor edge-sharpness, Gabor features, and local area integral invariant features, extracted from imaging-derived tumor segmentations, tumor edge-sharpness was most useful for predicting sonic hedgehog and group 4 tumors. Receiver operating characteristic analysis revealed superior performance of the double 10-fold cross-validation model for predicting sonic hedgehog, group 3, and group 4 tumors when using combined T1- and T2-weighted images (area under the curve = 0.79, 0.70, and 0.83, respectively). With the independent 3-dataset cross-validation strategy, select radiomic features were predictive of sonic hedgehog (area under the curve = 0.70-0.73) and group 4 (area under the curve = 0.76-0.80) medulloblastoma. CONCLUSIONS: This study provides proof-of-concept results for the application of radiomic and machine learning approaches to a multi-institutional dataset for the prediction of medulloblastoma subgroups.