Deep learning techniques have emerged in de-noising low-dose computed tomography (CT) images to avoid the potential health risks of high ionizing radiation dose on patients. Although these post-processing methods display high quality denoised images, the denoising performance still has the potential to improve. The primary purpose of this work was to determine and analyze the most effective and efficient hybrid loss function in deep learning (DL)-based denoising network. Objective functions in deep learning algorithms are the main keys for optimizing the parameters of a network and can affect the quality of the denoised image significantly. Hence, this work examined the various combinations of the most common objective functions in CT denoising networks, namely L1 loss, per-pixel loss, perceptual loss, and structural dissimilarity loss. Further, a hyperparameter learning algorithm was also introduced to find the best scalable factors of the loss functions in each hybrid loss function combination. For simplic-ity, RED-CNN was used in this study to easily demonstrate the performance of the losses during the denoising process. Based on this experiment, the balance between these loss function via the gradient-based optimization algorithm could help in the generalizability prediction of designing future CT denoising networks.

Download full-text PDF

Source
http://dx.doi.org/10.1109/EMBC48229.2022.9871380DOI Listing

Publication Analysis

Top Keywords

loss function
16
hybrid loss
12
deep learning
12
loss
9
gradient-based optimization
8
optimization algorithm
8
quality denoised
8
objective functions
8
denoising networks
8
denoising
6

Similar Publications

Aims/hypothesis: UK standard care for type 2 diabetes is structured diabetes education, with no effects on HbA, small, short-term effects on weight and low uptake. We evaluated whether remotely delivered tailored diabetes education combined with commercial behavioural weight management is cost-effective compared with current standard care in helping people with type 2 diabetes to lower their blood glucose, lose weight, achieve remission and improve cardiovascular risk factors.

Methods: We conducted a pragmatic, randomised, parallel two-group trial.

View Article and Find Full Text PDF

Melatonin antagonizes bone loss induced by mechanical unloading via IGF2BP1-dependent mA regulation.

Cell Mol Life Sci

January 2025

The Key Laboratory of Aerospace Medicine, Ministry of Education, Air Force Medical University, Xi'an, 710032, Shaanxi, China.

Disuse bone loss is prone to occur in individuals who lack mechanical stimulation due to prolonged spaceflight or extended bed rest, rendering them susceptible to fractures and placing an enormous burden on social care; nevertheless, the underlying molecular mechanisms of bone loss caused by mechanical unloading have not been fully elucidated. Numerous studies have focused on the epigenetic regulation of disuse bone loss; yet limited research has been conducted on the impact of RNA modification bone formation in response to mechanical unloading conditions. In this study, we discovered that mA reader IGF2BP1 was downregulated in both osteoblasts treated with 2D clinostat and bone tissue in HLU mice.

View Article and Find Full Text PDF

Inverse design via topology optimization has led to innovations in integrated photonics and offers a promising way for designing high-efficiency on-chip couplers with a minimal footprint. In this work, we exploit topology optimization to design a compact vertical coupler incorporating a bottom reflector, which achieves sub-decibel coupling efficiency on the 220-nm silicon-on-insulator platform. The final design of the vertical coupler yields a predicted coupling efficiency of -0.

View Article and Find Full Text PDF

Craniometaphyseal dysplasia (CMD), a rare craniotubular disorder, occurs in an autosomal dominant (AD) or autosomal recessive (AR) form. CMD is characterized by hyperostosis of craniofacial bones and metaphyseal flaring of long bones. Many patients with CMD suffer from neurological symptoms.

View Article and Find Full Text PDF

Objective: Meta-analysis of the clinical efficacy of plate and external fixator fixation in the treatment of AO-C type distal radius fractures.

Methods: PubMed, Embase, Cochrane Medical Library, Web of Science, CNKI, Wanfang, VIP and SinoMed databases were searched for all literature on randomized controlled clinical trials of AO-C distal radius fractures. The search time limits were from each database.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!