Prostate Segmentation with Encoder-Decoder Densely Connected Convolutional Network (Ed-Densenet)

Yuan, Yixuan; Qin, Wenjian; Guo, Xiaoqing; Buyyounouski, Mark K.; Hancock, Steve; Han, Bin; Liu, Xing

doi:10.1109/isbi.2019.8759498

Cited by 23 publications

(10 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main difference between the FCN and U-Net structures is that the FCN does not learn the mapping of the high-level features to the original input resolution in a step-by-step manner as it relies only on the feature extraction part of the network to make the final classification. Although both these structures are able to produce good results on different organ segmentation tasks [26], many works on prostate segmentation show success using U-Net as their base model [21][22][23][24][25]. Moreover the basic U-Net has been used successfully for the segmentation of different parts of the prostate [27].…”

Section: Structurementioning

confidence: 99%

“…Deep learning algorithms based on the convolutional neural network (CNN) such as the Fully Convolutional Network (FCN) [18], U-Net [19] and DenseNet [20] have achieved outstanding results in prostate segmentation tasks [21][22][23][24][25][26][27]. The CNN utilises a number of convolutional and pooling layers for extracting features automatically.…”

Section: Introductionmentioning

confidence: 99%

“…It is important to understand which components are beneficial for the deep learning model to perform prostate segmentation. However, many studies often present their works in the final form [21][22][23][24][25] or incorporate post-processing methods into their pipeline to obtain more accurate segmentation results [26,30,31]. As a result, the effect of each component on the overall performance is hard to distinguish and/or unknown.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Optimisation of 2D U-Net Model Components for Automatic Prostate Segmentation on MRI

et al. 2020

View full text Add to dashboard Cite

In this paper, we develop an optimised state-of-the-art 2D U-Net model by studying the effects of the individual deep learning model components in performing prostate segmentation. We found that for upsampling, the combination of interpolation and convolution is better than the use of transposed convolution. For combining feature maps in each convolution block, it is only beneficial if a skip connection with concatenation is used. With respect to pooling, average pooling is better than strided-convolution, max, RMS or L2 pooling. Introducing a batch normalisation layer before the activation layer gives further performance improvement. The optimisation is based on a private dataset as it has a fixed 2D resolution and voxel size for every image which mitigates the need of a resizing operation in the data preparation process. Non-enhancing data preprocessing was applied and five-fold cross-validation was used to evaluate the fully automatic segmentation approach. We show it outperforms the traditional methods that were previously applied on the private dataset, as well as outperforming other comparable state-of-the-art 2D models on the public dataset PROMISE12.

show abstract

Section: Structurementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimisation of 2D U-Net Model Components for Automatic Prostate Segmentation on MRI

et al. 2020

View full text Add to dashboard Cite

show abstract

“…DenseNet [46][47][48][49] was firstly proposed by Huang et al in 2017. It connects each layer with others in the way of feedforward.…”

Section: Densely Connected Network For Detecting Layersmentioning

confidence: 99%

“…boxes for Scale 3; (109, 114), (121, 153), and (169, 173) are the anchor boxes for Scale 2; and (232, 214), (241, 203), and (259, 271) are the anchor boxes for Scale 1. The sizes of the anchor boxes for the UCS-AOD dataset are as follows: (19,22), (23,29), (31,38), (49,52), (63, 86), (80, 92), (101, 124), (118, 147), (152, 167), (225, 201), (231, 212), and (268, 279). Among them, (19,22), (23,29), and (31,38) are the anchor boxes for Scale 4; (49, 52), (63, 86), and (80, 92) are the anchor boxes for Scale 3; (101, 124), (118, 147), and (152, 167) are the anchor boxes for Scale 2, and (225, 201), (231, 212), and (268, 279) are the anchor boxes for Scale 1.…”

Section: Anchor Boxes Of Our Modelmentioning

confidence: 99%

MRFF-YOLO: A Multi-Receptive Fields Fusion Network for Remote Sensing Target Detection

Xu¹,

Wu²

2020

Remote Sensing

View full text Add to dashboard Cite

High-altitude remote sensing target detection has problems related to its low precision and low detection rate. In order to enhance the performance of detecting remote sensing targets, a new YOLO (You Only Look Once)-V3-based algorithm was proposed. In our improved YOLO-V3, we introduced the concept of multi-receptive fields to enhance the performance of feature extraction. Therefore, the proposed model was termed Multi-Receptive Fields Fusion YOLO (MRFF-YOLO). In addition, to address the flaws of YOLO-V3 in detecting small targets, we increased the detection layers from three to four. Moreover, in order to avoid gradient fading, the structure of improved DenseNet was chosen in the detection layers. We compared our approach (MRFF-YOLO) with YOLO-V3 and other state-of-the-art target detection algorithms on an Remote Sensing Object Detection (RSOD) dataset and a dataset of Object Detection in Aerial Images (UCS-AOD). With a series of improvements, the mAP (mean average precision) of MRFF-YOLO increased from 77.10% to 88.33% in the RSOD dataset and increased from 75.67% to 90.76% in the UCS-AOD dataset. The leaking detection rates are also greatly reduced, especially for small targets. The experimental results showed that our approach achieved better performance than traditional YOLO-V3 and other state-of-the-art models for remote sensing target detection.

show abstract

Automatic intraprostatic lesion segmentation in multiparametric magnetic resonance images with proposed multiple branch UNet

Chen

et al. 2020

Medical Physics

Self Cite

View full text Add to dashboard Cite

Purpose Contouring intraprostatic lesions is a prerequisite for dose‐escalating these lesions in radiotherapy to improve the local cancer control. In this study, a deep learning‐based approach was developed for automatic intraprostatic lesion segmentation in multiparametric magnetic resonance imaging (mpMRI) images contributing to clinical practice. Methods Multiparametric magnetic resonance imaging images from 136 patient cases were collected from our institution, and all these cases contained suspicious lesions with Prostate Imaging Reporting and Data System (PI‐RADS) score ≥ 4. The contours of the lesion and prostate were manually created on axial T2‐weighted (T2W), apparent diffusion coefficient (ADC) and high b‐value diffusion‐weighted imaging (DWI) images to provide the ground truth data. Then a multiple branch UNet (MB‐UNet) was proposed for the segmentation of an indistinct target in multi‐modality MRI images. An encoder module was designed with three branches for the three MRI modalities separately, to fully extract the high‐level features provided by different MRI modalities; an input module was added by using three sub‐branches for three consecutive image slices, to consider the contour consistency among different image slices; deep supervision strategy was also integrated into the network to speed up the convergency of the network and improve the performance. The probability maps of the background, normal prostate and lesion were output by the network to generate the segmentation of the lesion, and the performance was evaluated using the dice similarity coefficient (DSC) as the main metric. Results A total of 162 lesions were contoured on 652 image slices, with 119 lesions in the peripheral zone, 38 in the transition zone, four in the central zone and one in the anterior fibromuscular stroma. All prostates were also contoured on 1,264 image slices. As for the segmentation of lesions in the testing set, MB‐UNet achieved a per case DSC of 0.6333, specificity of 0.9993, sensitivity of 0.7056; and global DSC of 0.7205, specificity of 0.9993, sensitivity of 0.7409. All the three deep learning strategies adopted in this study contributed to the performance promotion of the MB‐UNet. Missing the DWI modality would degrade the segmentation performance more markedly compared with the other two modalities. Conclusions A deep learning‐based approach with proposed MB‐UNet was developed to automatically segment suspicious lesions in mpMRI images. This study makes it feasible to adopt boosting intraprostatic lesions in clinical practice to achieve better outcomes.

show abstract

Prostate Segmentation with Encoder-Decoder Densely Connected Convolutional Network (Ed-Densenet)

Cited by 23 publications

References 8 publications

Optimisation of 2D U-Net Model Components for Automatic Prostate Segmentation on MRI

Optimisation of 2D U-Net Model Components for Automatic Prostate Segmentation on MRI

MRFF-YOLO: A Multi-Receptive Fields Fusion Network for Remote Sensing Target Detection

Automatic intraprostatic lesion segmentation in multiparametric magnetic resonance images with proposed multiple branch UNet

Contact Info

Product

Resources

About