Evolutionary convolutional neural network for efficient brain tumor segmentation and overall survival prediction
作者：
Highlights：
• Employ accumulated gradient normalization to compensate for our small batch size.
• Identify redundant filters of a U-Net-based network using the genetic algorithm.
• Remove trivial filters of the network to reduce its inference costs.
• Use the pruned network for brain tumor segmentation and overall survival prediction.
• Pruned network works as well as the original network but with less inference cost.
摘要
•Employ accumulated gradient normalization to compensate for our small batch size.•Identify redundant filters of a U-Net-based network using the genetic algorithm.•Remove trivial filters of the network to reduce its inference costs.•Use the pruned network for brain tumor segmentation and overall survival prediction.•Pruned network works as well as the original network but with less inference cost.
论文关键词：Deep learning,Genetic algorithm,Network compression
论文评审过程：Received 2 May 2022, Revised 22 August 2022, Accepted 5 October 2022, Available online 10 October 2022, Version of Record 18 October 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118996
