Volumetric ConvNets with Mixed Residual Connections for Automated Prostate Segmentation from 3D MR Images

By Lequan Yu, Xin Yang, Hao Chen, Jing Qin, P. Heng. 2017

358 citations recorded by the source corpus.

Automated prostate segmentation from 3D MR images is very challenging due to large variations of prostate shape and indistinct prostate boundaries. We propose a novel volumetric convolutional neural network (ConvNet) with mixed residual connections to cope with this challenging problem. Compared with previous methods, our volumetric ConvNet has two compelling advantages. First, it is implemented in a 3D manner and can fully exploit the 3D spatial contextual information of input data to perform efficient, precise and volume-to-volume prediction. Second and more important, the novel combination of residual connections (i.e., long and short) can greatly improve the training efficiency and discriminative capability of our network by enhancing the information propagation within the ConvNet both locally and globally. While the forward propagation of location information can improve the segmentation accuracy, the smooth backward propagation of gradient flow can accelerate the convergence speed and enhance the discrimination capability. Extensive experiments on the open MICCAI PROMISE12 challenge dataset corroborated the effectiveness of the proposed volumetric ConvNet with mixed residual connections. Our method ranked the first in the challenge, outperforming other competitors by a large margin with respect to most of evaluation metrics. The proposed volumetric ConvNet is general enough and can be easily extended to other medical image analysis tasks, especially ones with limited training data.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Scalable Neural Architecture Search for 3D Medical Image Segmentation (2019, methodology): “ropose a NAS framework for AutoML in designing neural networks especially for 3D medical image segmentation. 3D U-Net [1] has been popularly used for segmenting high-resolution 3D medical images (see [4,8,11]) since both semantic as well as spatial information can be eciently exploited through skip ”
Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation (2019, background): “With the help of convolutional neural networks, applications such as image classification [22, 34] ,object detection [28], and semantic segmentation [7, 40] are developing at an unprecedented speed nowadays.”
MRI Augmentation via Elastic Registration for Brain Lesions Segmentation (2017, background): “…was originally developed for neuronal structures segmentation in electron microscopic stacks, but during the last two years a lot of modiﬁcations for medical images segmentation were proposed including a 3D version of the network [6] and applications to various domains (e.g., prostate cancer [22]).”
MS-Net: Multi-Site Network for Improving Prostate Segmentation With Heterogeneous MRI Data (2020, methodology): “We adopt an adapted 2D Residual-UNet [2] as the segmentation network backbone, which achieves remarkable performance in the prostate segmentation problem.”
A review: Deep learning for medical image segmentation using multi-modality fusion (2019, background): “Motivated by the success of deep learning, researches in medical image field have also attempted to apply deep learning-based approaches to medical image segmentation in the brain [23] [54] [48], lung [34], pancreas [19] [59], prostate [74] and multi-organ [79] [69].”
Uncertainty-aware multi-view co-training for semi-supervised medical image segmentation and domain adaptation (2020, background): “Deep learning-based approaches have been widely adopted for this task and have led to state-of-the-art performance [40, 48, 64, 65].”
UNETR: Transformers for 3D Medical Image Segmentation (2021, methodology): “CNN-based Segmentation Networks : Since the introduction of the seminal U-Net [36], CNN-based networks have achieved state-of-the-art results on various 2D and 3D various medical image segmentation tasks [15, 54, 49, 17, 28].”
Deep convolutional neural networks for brain image analysis on magnetic resonance imaging: a review (2017, methodology): “[142] considered a 3D extension of the U-Net along with skip and/or residual connections for addressing the MICCAI Prostate MR Image Segmentation (PROMISE12).”
Boundary loss for highly unbalanced segmentation (2018, background): “Another common practice is to assign weights to the different classes, inversely proportional to the frequency of the corresponding labels (Brosch et al., 2015; Ronneberger et al., 2015; Kamnitsas et al., 2017; Long et al., 2015; Yu et al., 2017).”
Self-Distillation: Towards Efficient and Compact Neural Networks (2021, neutral): “D Eep convolutional neural networks have shown promising results in many applications such as image classification [1], [2], [3], [4], object detection [5], [6], [7], [8] and segmentation [9], [10], [11].”
A review on deep learning in medical image analysis (2021, background): “Top-down feedback for more accurate inference integrates ambiguous results [119, 120].”
Boundary-weighted Domain Adaptive Neural Network for Prostate MR Image Segmentation (2019, neutral): “[21] designed an efficient volumetric CNN”
PSNet: prostate segmentation on MRI based on a convolutional neural network (2018, background): “Yu et al.(22) proposed a volumetric ConvNet to segment prostate from MRI.”
Superhuman Accuracy on the SNEMI3D Connectomics Challenge (2017, methodology): “Residual variants of U-Net were previously applied to biomedical image segmentation [16, 17] and serial section EM image segmentation [9, 18].”