351. MRI features predict p53 status in lower-grade gliomas via a machine-learning approach
- Author
-
Yinyan Wang, Xing Fan, Yiming Li, Xing Liu, Kaibin Xu, Shaowu Li, Tao Jiang, Kai Wang, and Zenghui Qian
- Subjects
Male ,p53 ,Computer science ,Radiogenomics ,computer.software_genre ,lcsh:RC346-429 ,030218 nuclear medicine & medical imaging ,Correlation ,Machine Learning ,0302 clinical medicine ,Wavelet ,Image Processing, Computer-Assisted ,Child ,Brain Neoplasms ,Regular Article ,Glioma ,Middle Aged ,Magnetic Resonance Imaging ,Neurology ,030220 oncology & carcinogenesis ,Child, Preschool ,Lower-grade gliomas ,lcsh:R858-859.7 ,Female ,Adult ,Adolescent ,Cognitive Neuroscience ,Feature selection ,lcsh:Computer applications to medicine. Medical informatics ,Machine learning ,03 medical and health sciences ,Young Adult ,P53 status ,Entropy (information theory) ,Humans ,Radiology, Nuclear Medicine and imaging ,lcsh:Neurology. Diseases of the nervous system ,Aged ,Retrospective Studies ,Lower grade ,Receiver operating characteristic ,business.industry ,Pattern recognition ,ROC Curve ,Mutation ,Neurology (clinical) ,Artificial intelligence ,Tumor Suppressor Protein p53 ,business ,Prediction ,computer - Abstract
Background P53 mutation status is a pivotal biomarker for gliomas. Here, we developed a machine-learning model to predict p53 status in lower-grade gliomas based on radiomic features extracted from conventional magnetic resonance (MR) images. Methods Preoperative MR images were retrospectively obtained from 272 patients with primary grade II/III gliomas. The patients were randomly allocated in a 2:1 ratio to a training (n = 180) or validation (n = 92) set. A total of 431 radiomic features were extracted from each patient. The lest absolute shrinkage and selection operator (LASSO) method was used for feature selection and radiomic signature construction. Subsequently, a machine-learning model to predict p53 status was established using the selected features and a Support Vector Machine classifier. The predictive performance of all individual features and the model was calculated using receiver operating characteristic curves in both the training and validation sets. Results The p53-related radiomic signature was built using the LASSO algorithm; this procedure consisted of four first-order statistics or related wavelet features (including Maximum, Median, Minimum, and Uniformity), a shape and size-based feature (Spherical Disproportion), and ten textural features or related wavelet features (including Correlation, Run Percentage, and Sum Entropy). The prediction accuracies based on the area under the curve were 89.6% in the training set and 76.3% in the validation set, which were better than individual features. Conclusions These results demonstrate that MR image texture features are predictive of p53 mutation status in lower-grade gliomas. Thus, our procedure can be conveniently used to facilitate presurgical molecular pathological diagnosis., Highlights • We established a p53-related radiomic signature in lower-grade gliomas based on LASSO algorithm. • We developed a machine-learning model using the radiomic signature and a support vector machine. • P53 mutation status of lower-grade gliomas was predicted effectively based on our machine-learning model.
- Published
- 2017