Prediction of Brain Tumor Recurrence Location Based on Kullback–Leibler Divergence and Nonlinear Correlation Learning

Abstract

Brain tumor is one of the leading causes of cancer death. The high grade brain tumors are easier to recurrent even after standard treatment. Therefore, developing a method to predict brain tumor recurrence location plays an important role in the treatment planning and it can potentially prolong patient’s survival time. In this paper, we present a deep learning based brain tumor recurrence location prediction network. Since the dataset is usually small, we propose to use transfer learning to improve the prediction. We first train a multi-modal brain tumor segmentation network on public dataset BraTS 2018. Then, the pre-trained encoder is transferred to our private dataset to extract the semantic features. Following that, a multimodal fusion module and a nonlinear correlation learning module are designed to extract the effective features. To measure the similarity between the distributions of original features of one modality and the estimated correlated features of another modality, we propose to use Kullback-Leibler divergence. Based on this divergence, a correlation loss function is designed to maximize the similarity between the two feature distributions. Finally, two decoders are introduced to jointly segment the present brain tumor and predict its future tumor recurrence location. To the best of our knowledge, this is the first work that can segment the present tumor and at the same time predict future tumor recurrence location, making the treatment planning more efficient and precise. The experimental results demonstrated the effectiveness of our proposed method to predict the brain tumor recurrence sites from limited dataset.