Denoising of MR and CT images using cascaded multi-supervision convolutional neural networks with progressive training

Song, Hong; Chen, Lei; Cui, Yutao; Li, Qiang; Wang, Qi; Fan, Jingfan; Yang, Jian; Zhang, Le

doi:10.1016/j.neucom.2020.10.118

Cited by 30 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Initially, we employed the Spearman's correlation analysis to obtain the top 30 planar up/downregulated transition genes (Figure 3A (SR3 Table 1)) for each planar subtrajectory by inputting the planar global pseudospace-time trajectory (SR2 Figure 2) using stLearn software (12,(26)(27)(28)(29)(30)(31)(32)(33).…”

Section: The Trajectory-based Transition Gene Set For 3d Global Pseud...mentioning

confidence: 99%

Discovering hematoma-stimulated circuits for secondary brain injury after intraventricular hemorrhage by spatial transcriptome analysis

et al. 2023

Self Cite

View full text Add to dashboard Cite

IntroductionCentral nervous system (CNS) diseases, such as neurodegenerative disorders and brain diseases caused by acute injuries, are important, yet challenging to study due to disease lesion locations and other complexities.MethodsUtilizing the powerful method of spatial transcriptome analysis together with novel algorithms we developed for the study, we report here for the first time a 3D trajectory map of gene expression changes in the brain following acute neural injury using a mouse model of intraventricular hemorrhage (IVH). IVH is a common and representative complication after various acute brain injuries with severe mortality and mobility implications.ResultsOur data identified three main 3D global pseudospace-time trajectory bundles that represent the main neural circuits from the lateral ventricle to the hippocampus and primary cortex affected by experimental IVH stimulation. Further analysis indicated a rapid response in the primary cortex, as well as a direct and integrated effect on the hippocampus after IVH stimulation.DiscussionThese results are informative for understanding the pathophysiological changes, including the spatial and temporal patterns of gene expression changes, in IVH patients after acute brain injury, strategizing more effective clinical management regimens, and developing novel bioinformatics strategies for the study of other CNS diseases. The algorithm strategies used in this study are searchable via a web service (www.combio-lezhang.online/3dstivh/home).

show abstract

Section: The Trajectory-based Transition Gene Set For 3d Global Pseud...mentioning

confidence: 99%

Discovering hematoma-stimulated circuits for secondary brain injury after intraventricular hemorrhage by spatial transcriptome analysis

et al. 2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…SNR is linearly proportional to the main magnetic field (B0); hence, low-field MRI systems (< 1 T) inherently have significantly low SNR compared to conventional 1.5-3 T MRI scanners. Song et al [124] proposed a CNN-based auto encoder network with a transfer learning approach to learn a data-driven transformation from high-field noisy data with application to 0.35-T pelvic MR images. Tajima et al [125] studied the utility of a stacked U-Net method to reduce noise from the system.…”

Section: Noise In Anatomical Mrimentioning

confidence: 99%

Deep Learning for Image Enhancement and Correction in Magnetic Resonance Imaging—State-of-the-Art and Challenges

et al. 2022

View full text Add to dashboard Cite

Magnetic resonance imaging (MRI) provides excellent soft-tissue contrast for clinical diagnoses and research which underpin many recent breakthroughs in medicine and biology. The post-processing of reconstructed MR images is often automated for incorporation into MRI scanners by the manufacturers and increasingly plays a critical role in the final image quality for clinical reporting and interpretation. For image enhancement and correction, the post-processing steps include noise reduction, image artefact correction, and image resolution improvements. With the recent success of deep learning in many research fields, there is great potential to apply deep learning for MR image enhancement, and recent publications have demonstrated promising results. Motivated by the rapidly growing literature in this area, in this review paper, we provide a comprehensive overview of deep learning-based methods for post-processing MR images to enhance image quality and correct image artefacts. We aim to provide researchers in MRI or other research fields, including computer vision and image processing, a literature survey of deep learning approaches for MR image enhancement. We discuss the current limitations of the application of artificial intelligence in MRI and highlight possible directions for future developments. In the era of deep learning, we highlight the importance of a critical appraisal of the explanatory information provided and the generalizability of deep learning algorithms in medical imaging.

show abstract

“…Initially, we employed the Spearman correlation analysis of stLearn 9,[19][20][21][22][23][24][25][26] to obtain the top 30 planar upregulated/downregulated transition genes (Figure 3a and SR3. Table 1) for each planar subtrajectory by inputting the planar global pseudospace-time trajectory (SR2.…”

Section: The Trajectory-based Transition Gene Set For 3d Global Pseud...mentioning

confidence: 99%

Discovering Haematoma-Stimulated Circuits for Secondary Brain Injury after Intraventricular Haemorrhage by Spatial Transcriptome Analysis

Zhang

Badai

Wang

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Central nervous system (CNS) diseases, such as neurodegenerative disorders and brain diseases caused by acute injuries, are important yet challenging to study due to disease leisions' locations and other complexities. Utilizing the powerful spatial transcriptome analysis together with novel algorithms we developed for the study, we report here for the first time a 3D trajectory map of gene expression changes in the brain following acute neural injury using a mouse model of intraventricular haemorrhage (IVH). IVH is a common and representative complication after various acute brain injuries with severe mortality and mobility implications. Our data identified three main 3D global pseudospace-time trajectory bundles which represent the main neural circuits from the lateral ventricle to the hippocampus and primary cortex induced by experimental intraventricular haematoma stimulation. Further analysis indicated a rapid response in the primary cortex, as well as a direct and integrated effect on the hippocampus after IVH. These results are informative in understanding the pathophysiological changes, including the spatial and temporal patterns of gene expression changes, in IVH patients after acute brain injuries and strategizing more effective clinical management regimens. The bioinformatics strategies will also be useful for the study of other CNS diseases.

show abstract

Denoising of MR and CT images using cascaded multi-supervision convolutional neural networks with progressive training

Cited by 30 publications

References 25 publications

Discovering hematoma-stimulated circuits for secondary brain injury after intraventricular hemorrhage by spatial transcriptome analysis

Discovering hematoma-stimulated circuits for secondary brain injury after intraventricular hemorrhage by spatial transcriptome analysis

Deep Learning for Image Enhancement and Correction in Magnetic Resonance Imaging—State-of-the-Art and Challenges

Discovering Haematoma-Stimulated Circuits for Secondary Brain Injury after Intraventricular Haemorrhage by Spatial Transcriptome Analysis

Contact Info

Product

Resources

About