MT-UNET: A Novel U-Net Based Multi-Task Architecture For Visual Scene Understanding

Jha, Anand K.; Kumar, Awanish; Pande, Shivam; Banerjee, Biplab; Chaudhuri, Subhasis

doi:10.1109/icip40778.2020.9190695

Cited by 14 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MDWF-Net is a CNN [22,23] capable of calculating water-fat images, R2* and ∆f after receiving multi-echo GRE acquisitions as input. The architecture of MDWF-Net is based on multi-task U-Net, which has been previously proposed in the literature for signal processing tasks [18,19]. This configuration consists of an encoderdecoder structure that translates the input to a reduceddimensions latent space of features, from which water-fat images, R2* and ∆f are separately decodified (Fig.…”

Section: Multi-decoder Water-fat Separation Networkmentioning

confidence: 99%

Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes

et al. 2023

View full text Add to dashboard Cite

Objective To accurately estimate liver PDFF from chemical shift-encoded (CSE) MRI using a deep learning (DL)-based Multi-Decoder Water-Fat separation Network (MDWF-Net), that operates over complex-valued CSE-MR images with only 3 echoes. Methods The proposed MDWF-Net and a U-Net model were independently trained using the first 3 echoes of MRI data from 134 subjects, acquired with conventional 6-echoes abdomen protocol at 1.5 T. Resulting models were then evaluated using unseen CSE-MR images obtained from 14 subjects that were acquired with a 3-echoes CSE-MR pulse sequence with a shorter duration compared to the standard protocol. Resulting PDFF maps were qualitatively assessed by two radiologists, and quantitatively assessed at two corresponding liver ROIs, using Bland Altman and regression analysis for mean values, and ANOVA testing for standard deviation (STD) (significance level: .05). A 6-echo graph cut was considered ground truth. Results Assessment of radiologists demonstrated that, unlike U-Net, MDWF-Net had a similar quality to the ground truth, despite it considered half of the information. Regarding PDFF mean values at ROIs, MDWF-Net showed a better agreement with ground truth (regression slope = 0.94, R2 = 0.97) than U-Net (regression slope = 0.86, R2 = 0.93). Moreover, ANOVA post hoc analysis of STDs showed a statistical difference between graph cuts and U-Net (p < .05), unlike MDWF-Net (p = .53). Conclusion MDWF-Net showed a liver PDFF accuracy comparable to the reference graph cut method, using only 3 echoes and thus allowing a reduction in the acquisition times. Clinical relevance statement We have prospectively validated that the use of a multi-decoder convolutional neural network to estimate liver proton density fat fraction allows a significant reduction in MR scan time by reducing the number of echoes required by 50%. Key Points • Novel water-fat separation neural network allows for liver PDFF estimation by using multi-echo MR images with a reduced number of echoes. • Prospective single-center validation demonstrated that echo reduction leads to a significant shortening of the scan time, compared to standard 6-echo acquisition. • Qualitative and quantitative performance of the proposed method showed no significant differences in PDFF estimation with respect to the reference technique.

show abstract

Section: Multi-decoder Water-fat Separation Networkmentioning

confidence: 99%

Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In channel attention, the contribution of each channel in a given feature map is weighted by aggregating features spatially. Jha et al [78], for instance, apply task-specific channel-attention modules in order to highlight relevant task-specific features. In spatial attention, on the other hand, features are aggregated channel-wise, in order to provide the attention values for each spatial position in a given feature map.…”

Section: B Local and Global Context Modelingmentioning

confidence: 99%

A Threefold Review on Deep Semantic Segmentation: Efficiency-oriented, Temporal and Depth-aware design

Manfio¹,

Osório²

2023

Preprint

View full text Add to dashboard Cite

Semantic image and video segmentation stand among the most important tasks in computer vision nowadays, since they provide a complete and meaningful representation of the environment by means of a dense classification of the pixels in a given scene. Recently, Deep Learning, and more precisely Convolutional Neural Networks, have boosted semantic segmentation to a new level in terms of performance and generalization capabilities. However, designing Deep Semantic Segmentation models is a complex task, as it may involve application-dependent aspects. Particularly, when considering autonomous driving applications, the robustness-efficiency tradeoff, as well as intrinsic limitations -computational/memory bounds and data-scarcity -and constraints -real-time inference -should be taken into consideration. In this respect, the use of additional data modalities, such as depth perception for reasoning on the geometry of a scene, and temporal cues from videos to explore redundancy and consistency, are promising directions yet not explored to their full potential in the literature. In this paper, we conduct a survey on the most relevant and recent advances in Deep Semantic Segmentation in the context of vision for autonomous vehicles, from three different perspectives: efficiency-oriented model development for real-time operation, RGB-Depth data integration (RGB-D semantic segmentation), and the use of temporal information from videos in temporalaware models. Our main objective is to provide a comprehensive discussion on the main methods, advantages, limitations, results and challenges faced from each perspective, so that the reader can not only get started, but also be up to date in respect to recent advances in this exciting and challenging research field.

show abstract

“…Another approach, TransFuse (Zhang et al, 2021), employed a parallel combination of Transformers and CNNs to improve the efficiency of capturing global information. (Jha et al, 2020), UNet-2022 (Guo et al, 2022a), nnUNet (Isensee et al, 2021) methods for (a) Automated cardiac diagnosis (Bernard et al, 2018) and (b) Skin lesion segmentation (Gutman et al, 2016;Barata et al, 2014). The computational complexity of each method is reflected in the FLOPs(G) (Floating Point Operations) metric, while the segmentation performance is measured by the DSC(%) (Dice Similarity Coefficient).…”

Section: Medical Image Segmentationmentioning

confidence: 99%

Preface: Celebrating the 70th Anniversary of School of Mechanical Science and Engineering, Huazhong University of Science and Technology

Huang

Yin

2022

Sci. China Technol. Sci.

View full text Add to dashboard Cite

The shifted fractional trapezoidal rule (SFTR) with a special shift is adopted to construct a finite difference scheme for the time-fractional Allen-Cahn (tFAC) equation. Some essential key properties of the weights of SFTR are explored for the first time. Based on these properties, we rigorously demonstrate the discrete energy decay property and maximum-principle preservation for the scheme. Numerical investigations show that the scheme can resolve the intrinsic initial singularity of such nonlinear fractional equations as tFAC equation on uniform meshes without any correction. Comparison with the classic fractional BDF2 and L2-1 σ method further validates the superiority of SFTR in solving the tFAC equation. Experiments concerning both discrete energy decay and discrete maximum-principle also verify the correctness of the theoretical results.

show abstract

MT-UNET: A Novel U-Net Based Multi-Task Architecture For Visual Scene Understanding

Cited by 14 publications

References 6 publications

Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes

Liver PDFF estimation using a multi-decoder water-fat separation neural network with a reduced number of echoes

A Threefold Review on Deep Semantic Segmentation: Efficiency-oriented, Temporal and Depth-aware design

Preface: Celebrating the 70th Anniversary of School of Mechanical Science and Engineering, Huazhong University of Science and Technology

Contact Info

Product

Resources

About