An unsupervised PET image reconstruction method based on learnable gradient descent

CN116758177BActive Publication Date: 2026-07-07ZHEJIANG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZHEJIANG UNIV
Filing Date
2023-05-04
Publication Date
2026-07-07

AI Technical Summary

Technical Problem

Existing PET image reconstruction methods rely on high-quality labeled images, which leads to excessive patient radiation exposure and data collection time. At the same time, the generated model reconstruction lacks data interpretability and constraints from the physical properties of PET.

Method used

An unsupervised PET image reconstruction method based on learnable gradient descent is adopted. A regularization term is designed using a dual-domain unsupervised loss function and a convolutional neural network. The mathematical model of PET imaging is transformed into an optimization problem with a regularization term. A learnable gradient descent network LDAnet is constructed and trained in an unsupervised manner.

Benefits of technology

It achieves mathematical interpretability and deep neural network fitting capabilities for PET image reconstruction without the need for high-quality label images, reduces radiation exposure and data collection time, and improves reconstruction quality and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116758177B_ABST
    Figure CN116758177B_ABST
Patent Text Reader

Abstract

The application discloses an unsupervised PET image reconstruction method based on a learnable gradient descent, a regularization term is designed by using an L2,1 norm of a convolutional neural network, and the regularization term is subjected to a smoothing approximation process, so that an explicit solution of the regularization term gradient descent can be obtained. Based on this, the application designs an LDA network composed of multiple modules, which has strong prior learning ability and strong interpretability. At the same time, the application proposes a double-domain unsupervised loss composed of an image domain loss and a measurement domain loss, wherein the image domain loss is designed by using rotational invariance, and the measurement domain loss is subjected to a random noise data enhancement operation, thereby solving the problem that a large number of labeled images are required for training in mainstream methods, and thus the application has better clinical feasibility. The application can reconstruct a high-quality PET image from low-count Sinogram projection data, and excellent reconstruction effect is verified on clinical data.
Need to check novelty before this filing date? Find Prior Art