English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The image reconstruction process in medical imaging can be treated as solvingan inverse problem. The inverse problem is usually solved using time-consumingiterative algorithms with sparsity or other constraints. Recently, deep neuralnetwork (DNN)-based methods have been developed to accelerate theinverse-problem-solving process. However, these methods typically adoptsupervised learning scheme, which requires ground truths, or labels of thesolutions, for training. In many applications, it would be challenging or evenimpossible to obtain the ground truth, such as the tissue reflectivity functionin ultrasound beamforming. In this study, a general framework based onself-supervised learning (SSL) scheme is proposed to train a DNN to solve theinverse problems. In this way, the measurements can be used as both the inputsand the labels during the training of DNN. The proposed SSL method is appliedto four typical linear inverse problems for validation, i.e., plane waveultrasound and photoacoustic image reconstructions, compressed sensing-basedsynthetic transmit aperture dataset recovery and deconvolution in ultrasoundlocalization microscopy. Results show that, using the proposed framework, thetrained DNN can achieve improved reconstruction accuracy with reducedcomputational time, compared with conventional methods.

A General Framework for Inverse Problem Solving using Self-Supervised  Deep Learning: Validations in Ultrasound and Photoacoustic Image  Reconstruction