Tumor segmentation and survival prediction in glioma with deep learning

Sun, Li1; Zhang, Songtao1; Luo, Lin2

10.1007/978-3-030-11726-9_8

2019

16113349

4th International MICCAI Brainlesion Workshop, BrainLes 2018 held in conjunction with the Medical Image Computing for Computer Assisted Intervention Conference, MICCAI 2018

11384 LNCS

83-93

Granada, Spain

Springer Verlag

Every year, about 238,000 patients are diagnosed with brain tumor in the world. Accurate and robust tumor segmentation and prediction of patients’ overall survival are important for diagnosis, treatment planning and risk factor characterization. Here we present a deep learning-based framework for brain tumor segmentation and survival prediction in glioma using multimodal MRI scans. For tumor segmentation, we use ensembles of three different 3D CNN architectures for robust performance through majority rule. This approach can effectively reduce model bias and boost performance. For survival prediction, we extract 4524 radiomic features from segmented tumor region. Then decision tree and cross validation are used to select potent features. Finally, a random forest model is trained to predict the overall survival of patients. On 2018 MICCAI Multimodal Brain Tumor Segmentation Challenge (BraTS), our method ranks at second place and 5th place out of 60+ participating teams on survival prediction task and segmentation task respectively, achieving a promising 61.0% accuracy on classification of long-survivors, mid-survivors and short-survivors.
© Springer Nature Switzerland AG 2019.

第一

EI

20191306707783

Brain ; Decision trees ; Diagnosis ; Forecasting ; Medical imaging ; Patient treatment ; Tumors

Bioengineering and Biology:461 ; Systems Science:961

EV Compendex

1.Southern University of Science and Technology, Shenzhen; 518055, China
2.Peking University, Beijing; 100871, China

Sun, Li,Zhang, Songtao,Luo, Lin. Tumor segmentation and survival prediction in glioma with deep learning[C]:Springer Verlag,2019:83-93.