CT Liver Segmentation via PVT-based Encoding and Refined Decoding
CoRR(2024)
摘要
Accurate liver segmentation from CT scans is essential for computer-aided
diagnosis and treatment planning. Recently, Vision Transformers achieved a
competitive performance in computer vision tasks compared to convolutional
neural networks due to their exceptional ability to learn global
representations. However, they often struggle with scalability, memory
constraints, and computational inefficiency, particularly in handling
high-resolution medical images. To overcome scalability and efficiency issues,
we propose a novel deep learning approach, PVTFormer, that is
built upon a pretrained pyramid vision transformer (PVT v2) combined with
advanced residual upsampling and decoder block. By integrating a refined
feature channel approach with hierarchical decoding strategy, PVTFormer
generates high quality segmentation masks by enhancing semantic features.
Rigorous evaluation of the proposed method on Liver Tumor Segmentation
Benchmark (LiTS) 2017 demonstrates that our proposed architecture not only
achieves a high dice coefficient of 86.78%, mIoU of 78.46%, but also obtains
a low HD of 3.50. The results underscore PVTFormer's efficacy in setting a new
benchmark for state-of-the-art liver segmentation methods. The source code of
the proposed PVTFormer is available at
.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要