Polar representation-based cell nucleus segmentation in non-small cell lung cancer histopathological images

Wu, Xiao; Jiang, Yanyun; Yao, Zhigang; Zhou, Xiaoming; Lian, Jian; Zheng, Yuanjie

doi:10.1016/j.bspc.2021.103028

Cited by 7 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cell nuclei produced by the nuclei segmentation module are then constructed as a graph and fed into a GCN module for spatial feature extraction. To achieve this aim, we use a polar representation-based instance segmentation model ( Xiao et al, 2021 ) from our previous work to learn the segmentation of nuclei; this model leverages fully convolutional one-stage object detection and consists of a backbone network, feature pyramid network, and task-specific heads. Specifically, when we input an original image via the proposed network, the position of the cell center point and the distance of n ( n = 36) root rays can be obtained; then, the coordinates of these points on the contour are calculated according to the angle and length, connecting these points starting from 0°; and, finally, the regions within the connected regions are taken as the results of instance segmentation.…”

Section: Methodsmentioning

confidence: 99%

“…Specifically, to capture complex tumor microenvironment information and semantic information of entire image patches, we designed a model with two independent feature extraction branches as follows. 1) The GCN module, including a polar representation-based instance segmentation model ( Xiao et al, 2021 ), is used to extract all the cell nuclei contained in the histopathological patch and extract a nuclear feature composition map, which is used as an input to the GCN network to extract cell structural features. 2) The CNN module directly extracts semantic information from the whole patch to supplement the information loss of the GCN module.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

LAD-GCN: Automatic diagnostic framework for quantitative estimation of growth patterns during clinical evaluation of lung adenocarcinoma

Jiang

Yao

et al. 2022

Front. Physiol.

Self Cite

View full text Add to dashboard Cite

Quantitative estimation of growth patterns is important for diagnosis of lung adenocarcinoma and prediction of prognosis. However, the growth patterns of lung adenocarcinoma tissue are very dependent on the spatial organization of cells. Deep learning for lung tumor histopathological image analysis often uses convolutional neural networks to automatically extract features, ignoring this spatial relationship. In this paper, a novel fully automated framework is proposed for growth pattern evaluation in lung adenocarcinoma. Specifically, the proposed method uses graph convolutional networks to extract cell structural features; that is, cells are extracted and graph structures are constructed based on histopathological image data without graph structure. A deep neural network is then used to extract the global semantic features of histopathological images to complement the cell structural features obtained in the previous step. Finally, the structural features and semantic features are fused to achieve growth pattern prediction. Experimental studies on several datasets validate our design, demonstrating that methods based on the spatial organization of cells are appropriate for the analysis of growth patterns.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

LAD-GCN: Automatic diagnostic framework for quantitative estimation of growth patterns during clinical evaluation of lung adenocarcinoma

Jiang

Yao

et al. 2022

Front. Physiol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another drawback is the process of choosing the patches. A percentage threshold for the total number of pixels was used to determine whether a patch is questionable or not Year: 2021 Xiao et al ( 2021 ) proposed a Polar representation-based nucleus segmentation model in non-small lung cancer histopathological images Features: Backbone: Not mentioned Loss: Polar centeredness and polar IoU loss Stochastic Gradient Descent (SGD) was used as an optimizer. The proposed module employed centre classification and length regression to produce the contour of the nucleus in a polar coordinate, thus playing a dynamic role Comparison: U-Net, ExtremeNet, TensorMask, and PolarMask Datasets: They were manually collected 4792 histopathological slides with the lesions caused by non-small cell lung cancer from Shandong Provincial Hospital Parameters: F1-Score, Dice, Hausdorff and Aggregated Jaccard Index (AJI) Inference: The proposed model was better in terms of performance as compared to other models, thus indicating that the proposed approach could be a hypothetically valuable tool for clinical practices.…”

Section: Survey On Deep Learning Based Nucleus Segmentationmentioning

confidence: 99%

“…Xiao et al ( 2021 ) proposed a Polar representation-based nucleus segmentation model in non-small lung cancer histopathological images…”

Section: Survey On Deep Learning Based Nucleus Segmentationmentioning

confidence: 99%

A survey on recent trends in deep learning for nucleus segmentation from histopathology images

et al. 2023

View full text Add to dashboard Cite

Nucleus segmentation is an imperative step in the qualitative study of imaging datasets, considered as an intricate task in histopathology image analysis. Segmenting a nucleus is an important part of diagnosing, staging, and grading cancer, but overlapping regions make it hard to separate and tell apart independent nuclei. Deep Learning is swiftly paving its way in the arena of nucleus segmentation, attracting quite a few researchers with its numerous published research articles indicating its efficacy in the field. This paper presents a systematic survey on nucleus segmentation using deep learning in the last five years (2017–2021), highlighting various segmentation models (U-Net, SCPP-Net, Sharp U-Net, and LiverNet) and exploring their similarities, strengths, datasets utilized, and unfolding research areas.

show abstract

“…Xiao et al. ( 11 ) proposed a polar representation-based algorithm for non-small lung cancer segmentation from HIs. In cell nucleus segmentation, Pan et al.…”

Section: Introductionmentioning

confidence: 99%

GCLDNet: Gastric cancer lesion detection network combining level feature aggregation and attention feature fusion

Wang

Liu

et al. 2022

Front. Oncol.

View full text Add to dashboard Cite

BackgroundAnalysis of histopathological slices of gastric cancer is the gold standard for diagnosing gastric cancer, while manual identification is time-consuming and highly relies on the experience of pathologists. Artificial intelligence methods, particularly deep learning, can assist pathologists in finding cancerous tissues and realizing automated detection. However, due to the variety of shapes and sizes of gastric cancer lesions, as well as many interfering factors, GCHIs have a high level of complexity and difficulty in accurately finding the lesion region. Traditional deep learning methods cannot effectively extract discriminative features because of their simple decoding method so they cannot detect lesions accurately, and there is less research dedicated to detecting gastric cancer lesions.MethodsWe propose a gastric cancer lesion detection network (GCLDNet). At first, GCLDNet designs a level feature aggregation structure in decoder, which can effectively fuse deep and shallow features of GCHIs. Second, an attention feature fusion module is introduced to accurately locate the lesion area, which merges attention features of different scales and obtains rich discriminative information focusing on lesion. Finally, focal Tversky loss (FTL) is employed as a loss function to depress false-negative predictions and mine difficult samples.ResultsExperimental results on two GCHI datasets of SEED and BOT show that DSCs of the GCLDNet are 0.8265 and 0.8991, ACCs are 0.8827 and 0.8949, JIs are 0.7092 and 0.8182, and PREs are 0.7820 and 0.8763, respectively.ConclusionsExperimental results demonstrate the effectiveness of GCLDNet in the detection of gastric cancer lesions. Compared with other state-of-the-art (SOTA) detection methods, the GCLDNet obtains a more satisfactory performance. This research can provide good auxiliary support for pathologists in clinical diagnosis.

show abstract

Polar representation-based cell nucleus segmentation in non-small cell lung cancer histopathological images

Cited by 7 publications

References 24 publications

LAD-GCN: Automatic diagnostic framework for quantitative estimation of growth patterns during clinical evaluation of lung adenocarcinoma

LAD-GCN: Automatic diagnostic framework for quantitative estimation of growth patterns during clinical evaluation of lung adenocarcinoma

A survey on recent trends in deep learning for nucleus segmentation from histopathology images

GCLDNet: Gastric cancer lesion detection network combining level feature aggregation and attention feature fusion

Contact Info

Product

Resources

About