DilatedSegNet A Deep Dilated Segmentation Network for Polyp Segmentation Nikhil Kumar Tomar Debesh Jha Ulas Bagci

2025-05-03 0 0 825.18KB 11 页 10玖币

侵权投诉

DilatedSegNet: A Deep Dilated Segmentation

Network for Polyp Segmentation

Nikhil Kumar Tomar, Debesh Jha, Ulas Bagci

Machine & Hybrid Intelligence Lab, Department of Radiology,

Northwestern University

Abstract. Colorectal cancer (CRC) is the second leading cause of cancer-

related death worldwide. Excision of polyps during colonoscopy helps

reduce mortality and morbidity for CRC. Powered by deep learning,

computer-aided diagnosis (CAD) systems can detect regions in the colon

overlooked by physicians during colonoscopy. Lacking high accuracy and

real-time speed are the essential obstacles to be overcome for successful

clinical integration of such systems. While literature is focused on im-

proving accuracy, the speed parameter is often ignored. Toward this crit-

ical need, we intend to develop a novel real-time deep learning-based ar-

chitecture, DilatedSegNet, to perform polyp segmentation on the ﬂy. Di-

latedSegNet is an encoder-decoder network that uses pre-trained ResNet50

as the encoder from which we extract four levels of feature maps. Each

of these feature maps is passed through a dilated convolution pooling

(DCP) block. The outputs from the DCP blocks are concatenated and

passed through a series of four decoder blocks that predicts the segmen-

tation mask. The proposed method achieves a real-time operation speed

of 33.68 frames per second with an average dice coeﬃcient (DSC) of 0.90

and mIoU of 0.83. Additionally, we also provide heatmap along with

the qualitative results that shows the explanation for the polyp loca-

tion, which increases the trustworthiness of the method. The results on

the publicly available Kvasir-SEG and BKAI-IGH datasets suggest that

DilatedSegNet can give real-time feedback while retaining a high DSC,

indicating high potential for using such models in real clinical settings in

the near future. The GitHub link of the source code can be found here:

https://github.com/nikhilroxtomar/DilatedSegNet.

Keywords: Deep learning, polyp segmentation, colonoscopy, residual

network, generalizability, real-time segmentation

1 Introduction

Missed polyp during routine colonoscopy examination is the primary source of

interval colorectal cancer (CRC). The polyps that are not recognized within the

colonoscope are the major source contributor to this problem. Colonoscopy is

considered the gold standard for colon cancer diagnosis and follow-up. However,

22-28% of polyps are missed during a routine examination [12]. Some of these

polyps can cause post-colonoscopy colorectal cancer (CRC). One of the reasons

arXiv:2210.13595v1 [eess.IV] 24 Oct 2022

2 N. Tomar et al.

for the polyp miss-rate is either the polyp was not visible during the examination

or was not recognized despite being in the visual ﬁeld because of the faster

colonoscope withdrawal time. Deep learning based algorithms can highlight the

presence of pre-cancerous tissue in the colon and have the potential to improve

the diagnostic performance of endoscopists. Improving the polyp detection rate

as well as its accurate segmentation is an unmet clinical need. In practice, precise

polyp segmentation provides important information in the early detection of

colorectal cancer via their shape, texture, and location information.

Tomar et al. [17] proposed a feedback attention network for biomedical im-

age segmentation where they utilized the previous epoch mask with the current

training epoch in an iterative fashion to further improve the performance. Fan

et al. [3] used Res2Net-based [4] backbone where they used a parallel partial

decoder and parallel reverse attention mechanism for the accurate polyp seg-

mentation. Jha et al. [9] proposed an eﬃcient architecture where they utilized

the strength of the residual block, atrous spatial pyramidal pooling, with squeeze

and excitation block for polyp segmentation. Shen et al. [15] proposed a hard

region enhancement network (HRENet) that consists of an informative context

enhancement (ICE) module and trained the model on edge and structure con-

sistency aware loss (ESCLoss) to improve the polyp segmentation on the precise

edge. Zhao et al. [21] proposed a multi-scale subtraction network (MSNet) for

automatic polyp segmentation. Despite of several architectures proposed in the

literature, most existing methods often neglect the encoder and tend to focus

more on the decoder part of the network, which led to the loss of signiﬁcant

features from the encoder part. In our proposed method, we focus more on the

encoder part of the network by utilizing diﬀerent scales features which are passed

through multiple dilated convolutions to capture more enlarged features, leading

to improved polyp segmentation. Unlike other decoders, the design of our decoder

is straightforward. It utilizes simple sequences of layers such as an upsampling

layer, concatenation, residual block and an attention layer. We introduce the

novel deep learning architecture, DilatedSegNet, to address the critical need for

clinical integration of polyp segmentation routine, which is real-time and retains

high accuracy. The main contribution of the study are as follows:

1. We introduce a novel network named DilatedSegNet for polyp segmentation.

The architecture begins with a pre-trained ResNet50 [5] and utilizes dilated

convolution [19] pooling block to increase the receptive ﬁeld for capturing

more diverse and reliable features for a better delineation.

2. DilatedSegNet showed outstanding performance by outperforming nine stan-

dard benchmarking methods with two widely used publicly available polyp

segmentation datasets.

3. Extensive experimental results and cross-dataset test results on two unseen

datasets showed the better generalizability capability of the DilateSegNet.

Explored deep features showed via heatmaps that the proposed network

model is focusing on the target polyp regions and their boundaries, proving

visual interpretability of the model.

Title Suppressed Due to Excessive Length 3

Fig. 1: Block diagram of the proposed DilatedSegNet along with its components.

2 Method

Figure 3 shows the block diagram of the proposed DilatedSegNet along with its

core components. It follows an encoder-decoder scheme much like the U-Net [14],

consisting of a pre-trained ResNet50 [5] as an encoder. The input image Iwith

a resolution of [h×w×3] is fed to the pre-trained encoder from which we

extracts four levels of features maps {fi:i= 1,2,3,4}with varying resolution

of [h/2k×w/2k:k= 1,2,3,4]. Each of these feature maps is then passed

through a Dilated Convolution Pooling (DCP) block, where four parallel dilated

convolutions with the rate 1,3,6,9 are applied to enhance the ﬁeld of view.

The output from all the DCP blocks is concatenated and passed to the ﬁrst

decoder block, where the feature map is upsampled and concatenated with a skip

connection from the pre-trained encoder. Next, it is passed through some residual

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

DilatedSegNet:ADeepDilatedSegmentationNetworkforPolypSegmentationNikhilKumarTomar,DebeshJha,UlasBagciMachine&HybridIntelligenceLab,DepartmentofRadiology,NorthwesternUniversityAbstract.Colorectalcancer(CRC)isthesecondleadingcauseofcancer-relateddeathworldwide.Excisionofpolypsduringcolonoscopyhelpsred...

展开>> 收起<<

DilatedSegNet A Deep Dilated Segmentation Network for Polyp Segmentation Nikhil Kumar Tomar Debesh Jha Ulas Bagci.pdf

共11页,预览3页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

DilatedSegNet A Deep Dilated Segmentation Network for Polyp Segmentation Nikhil Kumar Tomar Debesh Jha Ulas Bagci

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: