UNet++: A Nested U-Net Architecture for Medical Image Segmentation

Zhou, Zongwei; Siddiquee, Mahfuzur Rahman; Tajbakhsh, Nima; Liang, Jianming

doi:10.48550/arxiv.1807.10165

Cited by 113 publications

(103 citation statements)

References 4 publications

Supporting

Mentioning

101

Contrasting

Unclassified

Order By: Relevance

“…Later works such as U-Net [18] improve upon this idea by employing a staged decoder with shortcuts to route highresolution feature maps directly to the decoder network. U-Net has proved to be very powerful, and there have been several U-Net based networks [19], [20], [21], [22] for specific tasks. Contrary to these approaches, work including those on context aggregation [23] advocate for maintaining high-resolution feature representations throughout.…”

Section: A Feature Learning In Cnns For Image Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Generalisable 3D Fabric Architecture for Streamlined Universal Multi-Dataset Medical Image Segmentation

Liu¹,

Dai²,

Engstrom³

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep neural networks are parameterised by weights that encode feature representations, whose performance is dictated through generalisation by using large-scale featurerich datasets. The lack of large-scale labelled 3D medical imaging datasets restrict constructing such generalised networks. In this work, a novel 3D segmentation network, Fabric Image Representation Encoding Network (FIRENet), is proposed to extract and encode generalisable feature representations from multiple medical image datasets in a large-scale manner. FIRENet learns image specific feature representations by way of 3D fabric network architecture that contains an exponential number of sub-architectures to handle various protocols and coverage of anatomical regions and structures. The fabric network uses Atrous Spatial Pyramid Pooling (ASPP) extended to 3D to extract local and image-level features at a fine selection of scales. The fabric is constructed with weighted edges allowing the learnt features to dynamically adapt to the training data at an architecture level. Conditional padding modules, which are integrated into the network to reinsert voxels discarded by feature pooling, allow the network to inherently process differentsize images at their original resolutions. FIRENet was trained for feature learning via automated semantic segmentation of pelvic structures and obtained a state-of-the-art median Dice Similarly Coefficient (DSC) score of 0.867. FIRENet was also simultaneously trained on Magnetic Resonance (MR) images acquired from 3D examinations of musculoskeletal elements in the (hip, knee, shoulder) joints and a public OAI knee dataset to perform automated segmentation of bone across anatomy. Transfer learning was used to show that the features learnt through the pelvic segmentation helped achieve improved mean DSC scores of 0.962, 0.963, 0.945 and 0.986 for automated segmentation of bone across datasets.

show abstract

Section: A Feature Learning In Cnns For Image Segmentationmentioning

confidence: 99%

“…Each block is a residual unit [8] with two convolutions followed by max-pooling. A convolutional layer is added to each encoderto-decoder shortcut to reduce semantic gaps [19].…”

Section: ) Encoder-decoder Backbonementioning

confidence: 99%

Generalisable 3D Fabric Architecture for Streamlined Universal Multi-Dataset Medical Image Segmentation

Liu¹,

Dai²,

Engstrom³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…U-Net [10] builds upon the idea of FCN and introduces a U-shape network with lateral connections between the contracting and expansive path which propagate context information to better localize. Since then, U-shape architecture thrives in many later works of 2D image segmentation [19,20,21,22,23] and 3D image segmentation [24,25]. U-Net++ [23] designs a more sophisticated structure with nested and dense lateral connection.…”

Section: Introductionmentioning

confidence: 99%

“…Since then, U-shape architecture thrives in many later works of 2D image segmentation [19,20,21,22,23] and 3D image segmentation [24,25]. U-Net++ [23] designs a more sophisticated structure with nested and dense lateral connection. By utilizing lateral connection at different level and nested upsample structure, U-Net++ manages to ensemble multiple U-Net to boost performance.…”

Section: Introductionmentioning

confidence: 99%

More than Encoder: Introducing Transformer Decoder to Upsample

Li¹,

Cai²,

Hu³

2021

Preprint

View full text Add to dashboard Cite

General segmentation models downsample images and then upsample to restore resolution for pixel level prediction. In such schema, upsample technique is vital in maintaining information for better performance. In this paper, we present a new upsample approach, Attention Upsample (AU), that could serve as general upsample method and be incorporated into any segmentation model that possesses lateral connections. AU leverages pixel-level attention to model long range dependency and global information for better reconstruction. It consists of Attention Decoder (AD) and bilinear upsample as residual connection to complement the upsampled features. AD adopts the idea of decoder from transformer which upsamples features conditioned on local and detailed information from contracting path. Moreover, considering the extensive memory and computation cost of pixel-level attention, we further propose to use window attention scheme to restrict attention computation in local windows instead of global range. Incorporating window attention, we denote our decoder as Window Attention Decoder (WAD) and our upsample method as Window Attention Upsample (WAU). We test our method on classic U-Net structure with lateral connection to deliver information from contracting path and achieve state-of-the-arts performance on Synapse (80.30 DSC and 23.12 HD) and MSD Brain (74.75 DSC) datasets.

show abstract

“…[14] optimized a SegNet [15] to segment dendrites of different alloys, including a 4D XCT. [7] identified Aluminides and Si phases in XCT using a U-Net, an architecture that, along with its many flavors [16][17][18][19], has shown success in a variety of applications [20][21][22]. Finally, [23] combined U-Nets with classic segmentation algorithms (e.g.…”

mentioning

confidence: 99%

A modular U-Net for automated segmentation of X-ray tomography images in composite materials

Bertoldo,

Decencière,

Ryckelynck

et al. 2021

Preprint

View full text Add to dashboard Cite

X-ray Computed Tomography (XCT) techniques have evolved to a point that high-resolution data can be acquired so fast that classic segmentation methods are prohibitively cumbersome, demanding automated data pipelines capable of dealing with non-trivial 3D images. Deep learning has demonstrated success in many image processing tasks, including material science applications, showing a promising alternative for a humanfree segmentation pipeline. In this paper a modular interpretation of U-Net (Modular U-Net) is proposed and trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66. We compare 2D and 3D versions of our model, finding that the former is slightly better than the latter. We observe that human-comparable results can be achievied even with only 10 annotated layers and using a shallow U-Net yields better results than a deeper one. As a consequence, Neural Network (NN) show indeed a promising venue to automate XCT data processing pipelines needing no human, adhoc intervention.

show abstract

UNet++: A Nested U-Net Architecture for Medical Image Segmentation

Cited by 113 publications

References 4 publications

Generalisable 3D Fabric Architecture for Streamlined Universal Multi-Dataset Medical Image Segmentation

Generalisable 3D Fabric Architecture for Streamlined Universal Multi-Dataset Medical Image Segmentation

More than Encoder: Introducing Transformer Decoder to Upsample

A modular U-Net for automated segmentation of X-ray tomography images in composite materials

Contact Info

Product

Resources

About