计算机科学
迭代重建
人工智能
正规化(语言学)
深度学习
降噪
人工神经网络
图像质量
特征(语言学)
操作员(生物学)
反问题
迭代法
模式识别(心理学)
忠诚
图像(数学)
算法
数学优化
数学
数学分析
哲学
语言学
生物化学
化学
抑制因子
转录因子
基因
电信
作者
Qiyang Zhang,Yingying Hu,Yixia Zhao,Jing Cheng,Wei Fan,Da Hu,Feng Shi,Shuangliang Cao,Yun Zhou,Yang Yongfeng,Xin Liu,Hairong Zheng,Dong Liang,Zhanli Hu
出处
期刊:IEEE Transactions on Medical Imaging
[Institute of Electrical and Electronics Engineers]
日期:2024-01-01
卷期号:43 (1): 122-134
被引量:3
标识
DOI:10.1109/tmi.2023.3293836
摘要
Low-count positron emission tomography (PET) imaging is challenging because of the ill-posedness of this inverse problem. Previous studies have demonstrated that deep learning (DL) holds promise for achieving improved low-count PET image quality. However, almost all data-driven DL methods suffer from fine structure degradation and blurring effects after denoising. Incorporating DL into the traditional iterative optimization model can effectively improve its image quality and recover fine structures, but little research has considered the full relaxation of the model, resulting in the performance of this hybrid model not being sufficiently exploited. In this paper, we propose a learning framework that deeply integrates DL and an alternating direction of multipliers method (ADMM)-based iterative optimization model. The innovative feature of this method is that we break the inherent forms of the fidelity operators and use neural networks to process them. The regularization term is deeply generalized. The proposed method is evaluated on simulated data and real data. Both the qualitative and quantitative results show that our proposed neural network method can outperform partial operator expansion-based neural network methods, neural network denoising methods and traditional methods.
科研通智能强力驱动
Strongly Powered by AbleSci AI