Hamrick, Forrest

Abstract Background Non-invasive differentiation between schwannomas and neurofibromas is important for appropriate management, preoperative counseling, and surgical planning, but has proven difficult using conventional imaging. The objective of this study was to develop and evaluate machine learning approaches for differentiating peripheral schwannomas from neurofibromas. Methods We assembled a cohort of schwannomas and neurofibromas from 3 independent institutions and extracted high-dimensional radiomic features from gadolinium-enhanced, T1-weighted MRI using the PyRadiomics package on Quantitative Imaging Feature Pipeline. Age, sex, neurogenetic syndrome, spontaneous pain, and motor deficit were recorded. We evaluated the performance of 6 radiomics-based classifier models with and without clinical features and compared model performance against human expert evaluators. Results One hundred and seven schwannomas and 59 neurofibromas were included. The primary models included both clinical and imaging data. The accuracy of the human evaluators (0.765) did not significantly exceed the no-information rate (NIR), whereas the Support Vector Machine (0.929), Logistic Regression (0.929), and Random Forest (0.905) classifiers exceeded the NIR. Using the method of DeLong, the AUCs for the Logistic Regression (AUC = 0.923) and K Nearest Neighbor (AUC = 0.923) classifiers were significantly greater than the human evaluators (AUC = 0.766; p = 0.041). Conclusions The radiomics-based classifiers developed here proved to be more accurate and had a higher AUC on the ROC curve than expert human evaluators. This demonstrates that radiomics using routine MRI sequences and clinical features can aid in differentiation of peripheral schwannomas and neurofibromas.

Abstract BACKGROUND Clinicoradiologic differentiation between benign and malignant peripheral nerve sheath tumors (PNSTs) has important management implications. OBJECTIVE To develop and evaluate machine-learning approaches to differentiate benign from malignant PNSTs. METHODS We identified PNSTs treated at 3 institutions and extracted high-dimensional radiomics features from gadolinium-enhanced, T1-weighted magnetic resonance imaging (MRI) sequences. Training and test sets were selected randomly in a 70:30 ratio. A total of 900 image features were automatically extracted using the PyRadiomics package from Quantitative Imaging Feature Pipeline. Clinical data including age, sex, neurogenetic syndrome presence, spontaneous pain, and motor deficit were also incorporated. Features were selected using sparse regression analysis and retained features were further refined by gradient boost modeling to optimize the area under the curve (AUC) for diagnosis. We evaluated the performance of radiomics-based classifiers with and without clinical features and compared performance against human readers. RESULTS A total of 95 malignant and 171 benign PNSTs were included. The final classifier model included 21 imaging and clinical features. Sensitivity, specificity, and AUC of 0.676, 0.882, and 0.845, respectively, were achieved on the test set. Using imaging and clinical features, human experts collectively achieved sensitivity, specificity, and AUC of 0.786, 0.431, and 0.624, respectively. The AUC of the classifier was statistically better than expert humans ( P = .002). Expert humans were not statistically better than the no-information rate, whereas the classifier was ( P = .001). CONCLUSION Radiomics-based machine learning using routine MRI sequences and clinical features can aid in evaluation of PNSTs. Further improvement may be achieved by incorporating additional imaging sequences and clinical variables into future models.

OBJECTIVE The aim of this study was to examine the role of intraoperative neuromonitoring (IONM) during resection of benign peripheral nerve sheath tumors in achieving gross-total resection (GTR) and in reducing postoperative neurological complications. METHODS Data from consecutive adult patients who underwent resection of a benign peripheral nerve sheath tumor at 7 participating institutions were combined. Propensity score matching was used to balance covariates. The primary outcomes of interest were the association between IONM and GTR and the association of IONM and the development of a permanent postoperative neurological complication. The secondary outcomes of interest were the association between IONM and GTR and the association between IONM and the development of a permanent postoperative neurological complication in the subgroup of patients with tumors involving a motor or mixed nerve. Univariate and multivariate logistic regression were then performed on the propensity score–matched samples to assess the ability of the independent variables to predict the outcomes of interest. RESULTS A total of 337 patients who underwent resection of benign nerve sheath tumors were included. In multivariate analysis, the use of IONM (OR 0.460, 95% CI 0.199–0.978; p = 0.047) was a significant negative predictor of GTR, whereas none of the variables, including IONM, were associated with the occurrence of a permanent postoperative neurological complication. Within the subgroup of motor/mixed nerve tumors, in the multivariate analysis, IONM (OR 0.263, 95% CI 0.096–0.723; p = 0.010) was a significant negative predictor of a GTR, whereas IONM (OR 3.800, 95% CI 1.925–7.502; p < 0.001) was a significant positive predictor of a permanent postoperative motor deficit. CONCLUSIONS Overall, 12% of the cohort had a permanent neurological complication, with new or worsened paresthesias most common, followed by pain and then weakness. The authors found that formal IONM was associated with a reduced likelihood of GTR and had no association with neurological complications. The authors believe that these data argue against IONM being considered standard of care but do not believe that these data should be used to universally argue against IONM during resection of benign nerve sheath tumors.