Survival prediction using ensemble tumor segmentation and transfer learning
This addresses survival prediction for brain tumor patients using imaging and clinical data, but it appears incremental as it builds on existing segmentation and transfer learning methods.
The paper tackles survival prediction for brain tumor patients by developing a cascaded pipeline that segments tumors and their subregions, then uses these results along with clinical and image features from a pretrained VGG-16 network. Preliminary results show promising segmentation performance, though survival prediction values need improvement with further testing on feature extraction.
Segmenting tumors and their subregions is a challenging task as demonstrated by the annual BraTS challenge. Moreover, predicting the survival of the patient using mainly imaging features, while being a desirable outcome to evaluate the treatment of the patient, it is also a difficult task. In this paper, we present a cascaded pipeline to segment the tumor and its subregions and then we use these results and other clinical features together with image features coming from a pretrained VGG-16 network to predict the survival of the patient. Preliminary results with the training and validation dataset show a promising start in terms of segmentation, while the prediction values could be improved with further testing on the feature extraction part of the network.