KiU-Net: Overcomplete Convolutional Architectures for Biomedical Image and Volumetric Segmentation (2020-10-04T00:00:00.000000Z)

TL;DR

A new architecture for im- age segmentation- KiU-Net is designed which has two branches: an overcomplete convolutional network Kite-Net which learns to capture fine details and accurate edges of the input, and (2) U- net which learns high level features.

Abstract

Most methods for medical image segmentation use U-Net or its variants as they have been successful in most of the applications. After a detailed analysis of these “traditional” encoder-decoder based approaches, we observed that they perform poorly in detecting smaller structures and are unable to segment boundary regions precisely. This issue can be attributed to the increase in receptive field size as we go deeper into the encoder. The extra focus on learning high level features causes U-Net based approaches to learn less information about low-level features which are crucial for detecting small structures. To overcome this issue, we propose using an overcomplete convolutional architecture where we project the input image into a higher dimension such that we constrain the receptive field from increasing in the deep layers of the network. We design a new architecture for im- age segmentation- KiU-Net which has two branches: (1) an overcomplete convolutional network Kite-Net which learns to capture fine details and accurate edges of the input, and (2) U-Net which learns high level features. Furthermore, we also propose KiU-Net 3D which is a 3D convolutional architecture for volumetric segmentation. We perform a detailed study of KiU-Net by performing experiments on five different datasets covering various image modalities. We achieve a good performance with an additional benefit of fewer parameters and faster convergence. We also demonstrate that the extensions of KiU-Net based on residual blocks and dense blocks result in further performance improvements. Code: https://github.com/jeya-maria-jose/KiU-Net-pytorch

Authors

References64 items

KiU-Net: Towards Accurate Segmentation of Biomedical Images using Over-complete Representations

UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation

CPFNet: Context Pyramid Fusion Network for Medical Image Segmentation

CAI-UNet for segmentation of liver lesion in CT image

MDAN-UNet: Multi-Scale and Dual Attention Enhanced Nested U-Net Architecture for Segmentation of Optical Coherence Tomography Images

Learning to Segment Brain Anatomy From 2D Ultrasound With Less Data

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

MC-Unet: Multi-scale Convolution Unet for Bladder Cancer Cell Segmentation in Phase-Contrast Microscopy Images

Brain Tumor Segmentation and Survival Prediction Using 3D Attention UNet

Automatic Segmentation of Brain Tumor from 3D MR Images Using SegNet, U-Net, and PSP-Net

Brain Tumor Segmentation Based on 3D Residual U-Net

Improving Nuclei/Gland Instance Segmentation in Histopathology Images by Full Resolution Neural Network and Spatial Constrained Loss

Automatic Couinaud Segmentation from CT Volumes on Liver Using GLC-UNet

Automatic Retinal Blood Vessel Segmentation Based on Fully Convolutional Neural Networks

Light-Weight Hybrid Convolutional Network for Liver Tumor Segmentation

Multilevel and Multiscale Deep Neural Network for Retinal Blood Vessel Segmentation

Deep Learning Techniques for Medical Image Segmentation: Achievements and Challenges

RIC-Unet: An Improved Neural Network Based on Unet for Nuclei Segmentation in Histology Images

The Liver Tumor Segmentation Benchmark (LiTS)

U-Net: deep learning for cell counting, detection, and morphometry

Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge

RA-UNet: A Hybrid Deep Attention-Aware Network to Extract Liver and Tumor in CT Scans

Weighted Res-UNet for High-Quality Retina Vessel Segmentation

Automatic Segmentation of the Cerebral Ventricle in Neonates Using Deep Learning with 3D Reconstructed Freehand Ultrasound Imaging

No New-Net

Glioma Prognosis: Segmentation of the Tumor and Survival Prediction Using Shape, Geometric and Clinical Information

Deep Learning Radiomics Algorithm for Gliomas (DRAG) Model: A Novel Approach Using 3D UNET Based Deep Convolutional Neural Network for Predicting Survival in Gliomas

3D MRI brain tumor segmentation using autoencoder regularization

S3D-UNet: Separable 3D U-Net for Brain Tumor Segmentation

Deep Convolutional Neural Networks Using U-Net for Automatic Brain Tumor Segmentation in Multimodal MRI Volumes

Recalibrating Fully Convolutional Networks With Spatial and Channel “Squeeze and Excitation” Blocks

Automatic Segmentation of Kidney and Renal Tumor in CT Images Based on 3D Fully Convolutional Neural Network with Pyramid Pooling Module

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

H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes

Automatic real-time CNN-based neonatal brain ventricles segmentation

Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation

Brain Tumor Segmentation Using a 3D FCN with Multi-scale Loss

Patch-based fully convolutional neural network with skip connections for retinal blood vessel segmentation

Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features

DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentation

MIMO-Net: A multi-input multi-output convolutional neural network for cell segmentation in fluorescence microscopy images

PixelNet: Representation of the pixels, by the pixels, and for the pixels

A survey on deep learning in medical image analysis

DCAN: Deep contour‐aware networks for object instance segmentation from histology images

Pyramid Scene Parsing Network

Image-to-Image Translation with Conditional Adversarial Networks

Topology Aware Fully Convolutional Networks for Histology Gland Segmentation

Densely Connected Convolutional Networks

3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Gland segmentation in colon histology images: The glas challenge contest

Deep Residual Learning for Image Recognition

SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

U-Net: Convolutional Networks for Biomedical Image Segmentation

Automated Separation of Binary Overlapping Trees in Low-Contrast Color Retinal Images

Over-complete representations on recurrent neural networks can support persistent percepts

Extracting and composing robust features with denoising autoencoders

User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability

Ridge-based vessel segmentation in color images of the retina

Learning Overcomplete Representations

3D Unet-based Kidney and Kidney Tumer Segmentation with Attentive Feature Learning

BiSC-UNet: A fine segmentation framework for kidney and renal tumor