SegT: A Novel Separated Edge-guidance Transformer Network for Polyp Segmentation

Feiyu Chen; Haiping Ma; Weijia Zhang

RECORD

SegT: A Novel Separated Edge-guidance Transformer Network for Polyp Segmentation

Feiyu Chen, Haiping Ma, Weijia Zhang

Computer Vision and Pattern Recognition · Submitted 19 Jun 2023

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim To evaluate the effectiveness of SegT, we conducted experiments with five challenging public datasets, and the proposed model achieved state-of-the-art performance.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2306.10773 (version 1)

Pith IDOKNWP6N5, Pith's permanent record of this paper and its review

CitationFeiyu Chen, Haiping Ma, Weijia Zhang. "SegT: A Novel Separated Edge-guidance Transformer Network for Polyp Segmentation." arXiv:2306.10773 (2023). https://pith.science/paper/OKNWP6N5

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 8,044 input tokens.

What the abstract itself states, as probabilities: claims first 0.15 · states limitation 0.09 · human participants 0.24 · controlled comparison 0.24 · randomized assignment 0.01 · states quantitative result 0.12 · compares with prior results 0.79 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

Accurate segmentation of colonoscopic polyps is considered a fundamental step in medical image analysis and surgical interventions. Many recent studies have made improvements based on the encoder-decoder framework, which can effectively segment diverse polyps. Such improvements mainly aim to enhance local features by using global features and applying attention methods. However, relying only on the global information of the final encoder block can result in losing local regional features in the intermediate layer. In addition, determining the edges between benign regions and polyps could be a challenging task. To address the aforementioned issues, we propose a novel separated edge-guidance transformer (SegT) network that aims to build an effective polyp segmentation model. A transformer encoder that learns a more robust representation than existing CNN-based approaches was specifically applied. To determine the precise segmentation of polyps, we utilize a separated edge-guidance module consisting of separator and edge-guidance blocks. The separator block is a two-stream operator to highlight edges between the background and foreground, whereas the edge-guidance block lies behind both streams to strengthen the understanding of the edge. Lastly, an innovative cascade fusion module was used and fused the refined multi-level features. To evaluate the effectiveness of SegT, we conducted experiments with five challenging public datasets, and the proposed model achieved state-of-the-art performance.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	Accurate segmentation of colonoscopic polyps is considered a fundamental step in medical image analysis and surgical interventions.	background 1.00	0.04	0.02	asserted 0.99
s1	Many recent studies have made improvements based on the encoder-decoder framework, which can effectively segment diverse polyps.	background 1.00	0.04	0.02	asserted 1.00
s2	Such improvements mainly aim to enhance local features by using global features and applying attention methods.	background 1.00	0.05	0.02	asserted 0.98
s3	However, relying only on the global information of the final encoder block can result in losing local regional features in the intermediate layer.	gap 0.82	0.09	0.02	asserted 0.59
s4	In addition, determining the edges between benign regions and polyps could be a challenging task.	gap 0.56	0.06	0.02	hedged possibility 1.00
s5	To address the aforementioned issues, we propose a novel separated edge-guidance transformer (SegT) network that aims to build an effective polyp segmentation model.	contribution 1.00	0.83	0.09	asserted 0.96
s6	A transformer encoder that learns a more robust representation than existing CNN-based approaches was specifically applied.	method 0.84	0.75	0.04	asserted 0.99
s7	To determine the precise segmentation of polyps, we utilize a separated edge-guidance module consisting of separator and edge-guidance blocks.	method 0.90	0.90	0.05	asserted 1.00
s8	The separator block is a two-stream operator to highlight edges between the background and foreground, whereas the edge-guidance block lies behind both streams to strengthen the understanding of the edge.	method 0.98	0.74	0.03	asserted 1.00
s9	Lastly, an innovative cascade fusion module was used and fused the refined multi-level features.	method 0.92	0.89	0.05	asserted 0.99
s10	To evaluate the effectiveness of SegT, we conducted experiments with five challenging public datasets, and the proposed model achieved state-of-the-art performance.	result 1.00	0.97	0.83	asserted 1.00

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.96.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

Accurate segmentation of colonoscopic polyps is considered a fundamental step in medical image analysis and surgical interventions. Many recent studies have made improvements based on the encoder-decoder framework, which can effectively segment diverse polyps. Such improvements mainly aim to enhance local features by using global features and applying attention methods. However, relying only on the global information of the final encoder block can result in losing local regional features in the intermediate layer. In addition, determining the edges between benign regions and polyps could be a challenging task. To address the aforementioned issues, we propose a novel separated edge-guidance transformer (SegT) network that aims to build an effective polyp segmentation model. A transformer encoder that learns a more robust representation than existing CNN-based approaches was specifically applied. To determine the precise segmentation of polyps, we utilize a separated edge-guidance module consisting of separator and edge-guidance blocks. The separator block is a two-stream operator to highlight edges between the background and foreground, whereas the edge-guidance block lies behind both streams to strengthen the understanding of the edge. Lastly, an innovative cascade fusion module was used and fused the refined multi-level features. To evaluate the effectiveness of SegT, we conducted experiments with five challenging public datasets, and the proposed model achieved state-of-the-art performance.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

1 arXiv paper cite this in 2 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it

Compares with it 1 paper

In addition, we also compare with Transformer-based methods, including TransFuse [ref] , HarDNet-MSEG [ref] , DS-TransUNet [ref] , Polyp-PVT [ref] , and SwinE-Net [ref] , and refinement-based methods, such as CaraNet [ref] , ColonFormer [ref] , and SegT [ref] .

Polyper: Boundary Sensitive Polyp Segmentation · Analysis of Experimental Results · 1.00

Cites it as background 1 paper

As an early attempt, SegT [ref] highlights the boundaries of the polyp area by assessing the disparity between the foreground and background regions.

Polyper: Boundary Sensitive Polyp Segmentation · Introduction · 0.91

What it cites · 4 sentences without a confident label

Uses the method or tool of 4 papers

Inspired by [ref] , we combine the two loss functions as the total loss for supervision with the following formula: [equation] where $L_wIOU$ and $L_wbce$ denote the weighted IoU loss and BCE loss for global and local restrictions, respectively.

Towards Accurate Camouflaged Object Detection with Mixture Convolution and Interactive Fusion · Loss function · 0.81
Considering the differences in the sizes of each polyp image, the input image is simply resized to 352 × 352, and we adopt a multi-scale training strategy thereafter [ref] .

A Competitive Method to VIPriors Object Detection Challenge · Implementation Details · 0.96
Considering the differences in the sizes of each polyp image, the input image is simply resized to 352 × 352, and we adopt a multi-scale training strategy thereafter [ref] .

HarDNet-MSEG: A Simple Encoder-Decoder Polyp Segmentation Neural Network that Achieves over 0.9 Mean Dice and 86 FPS · Implementation Details · 0.96

Compares with 1 paper

In Table [ref], we also report the results of three transformer-based methods (TransUNet [ref] , TransFuse [ref] , and Polyp-PVT [ref] ) with our proposed framework.

TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation · Result · 1.00

Cites as background 2 papers

In medical image segmentation, the TransUNet [ref] and TransFuse [ref] models are developed based on a transformer for polyp segmentation and have achieved good results.

TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation · Vision Transformer · 0.92
To apply transformers to computer vision tasks, Dosovitskiy et al. [ref] proposed a vision transformer (ViT), which is the first pure transformer for image classification.

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale · Vision Transformer · 0.92

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.