Polyp Segmentation with Fully Convolutional Deep Neural Networks—Extended Evaluation Study

Guo, Yun; Bernal, Jorge; Matuszewski, Bogdan J.

doi:10.3390/jimaging6070069

Cited by 41 publications

(18 citation statements)

References 61 publications

(81 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Later, the same group proposed the DeepLabV3+ [25] architecture that used skip connections between the encoder and decoder. Both of these networks have been widely used by the biomedical imaging community [26]- [28].…”

Section: Related Work a Medical Image Segmentationmentioning

confidence: 99%

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

Srivastava

Jha

Chanda

et al. 2022

IEEE J. Biomed. Health Inform.

135

View full text Add to dashboard Cite

Methods based on convolutional neural networks have improved the performance of biomedical image segmentation. However, most of these methods cannot efficiently segment objects of variable sizes and train on small and biased datasets, which are common for biomedical use cases. While methods exist that incorporate multi-scale fusion approaches to address the challenges arising with variable sizes, they usually use complex models that are more suitable for general semantic segmentation problems. In this paper, we propose a novel architecture called Multi-Scale Residual Fusion Network (MSRF-Net), which is specially designed for medical image segmentation. The proposed MSRF-Net is able to exchange multi-scale features of varying receptive fields using a Dual-Scale Dense Fusion (DSDF) block. Our DSDF block can exchange information rigorously across two different resolution scales, and our MSRF sub-network uses multiple DSDF blocks in sequence to perform multi-scale fusion. This allows the preservation of resolution, improved information flow and propagation of both high-and low-level features to obtain accurate segmentation maps. The proposed MSRF-Net allows to capture object variabilities and provides improved results on different biomedical datasets. Extensive experiments on MSRF-Net demonstrate that the proposed method outperforms the cutting-edge medical image segmentation methods on four publicly available datasets. We achieve the Dice Coefficient (DSC) of 0.9217, 0.9420, and 0.9224, 0.8824 on Kvasir-SEG, CVC-ClinicDB, 2018 Data Science Bowl dataset, and ISIC-2018 skin lesion segmentation challenge dataset respectively. We further conducted generalizability tests and achieved DSC of 0.7921 and 0.7575 on CVC-ClinicDB and Kvasir-SEG, respectively.

show abstract

Section: Related Work a Medical Image Segmentationmentioning

confidence: 99%

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

Srivastava

Jha

Chanda

et al. 2022

IEEE J. Biomed. Health Inform.

135

View full text Add to dashboard Cite

show abstract

“…Much progress has been made in histopathological imaging, but there are not many studies on the classification of colorectal cancer imaging. Furthermore, many results of various studies still have unreliable accuracy results [19], [20]. In this paper, we aim to address the gap in colorectal cancer imaging in the existing review of literature.…”

Section: Introductionmentioning

confidence: 99%

Colorectal multi-class image classification using deep learning models

Raju¹,

Rao²

2022

Bulletin EEI

View full text Add to dashboard Cite

Colorectal image classification is a novel application area in medical image processing. Colorectal images are one of the most prevalent malignant tumour disease type in the world. However, due to the complexity of histopathological imaging, the most accurate and effective classification still needs to be addressed. In this work we proposed a novel architecture of convolution neural network with deep learning models for the multiclass classification of histopathology images. We achieved the findings using three deep learning models, including the vgg16 with 96.16% and a modified version of Resnet50 with 97.08%, however the proposed Adaptive Resnet152 model generated the best accuracy of 98.38%. The colorectal image multiclass dataset is publicly available which has 5000 images with 8 classes. In this study we have increased all classes equally, total 15000 images have been generated using image augmentation technique. This dataset consists of 60% training images and 40% testing images. The suggested method in this paper produced better results than the existing histopathology image categorization methods with the lowest error rate. For histopathological image categorization, it is a straightforward, effective, and efficient method. We were able to attain state-of-the-art outcomes by efficiently utilizing the resourced dataset.

show abstract

“…The main variation across different works is the design of such architectures. For instance in [14] an encoder-decoder network containing multi-resolution, multi-classification, and fusion sub-networks was proposed, and in [32] several combinations of different encoder and decoder architectures were explored. In [12] an architecture in which there is a shared encoder and two mutually depending decoders that model polyp areas and boundaries respectively is introduced, whereas in [22] ensembles of instance-segmentation ar-Fig.…”

Section: Introductionmentioning

confidence: 99%

Double Encoder-Decoder Networks for Gastrointestinal Polyp Segmentation

Galdrán¹,

Carneiro²,

Ballester³

2021

Pattern Recognition. ICPR International Workshops and Challenges

View full text Add to dashboard Cite

Polyps represent an early sign of the development of Colorectal Cancer. The standard procedure for their detection consists of colonoscopic examination of the gastrointestinal tract. However, the wide range of polyp shapes and visual appearances, as well as the reduced quality of this image modality, turn their automatic identification and segmentation with computational tools into a challenging computer vision task. In this work, we present a new strategy for the delineation of gastrointestinal polyps from endoscopic images based on a direct extension of common encoder-decoder networks for semantic segmentation. In our approach, two pretrained encoder-decoder networks are sequentially stacked: the second network takes as input the concatenation of the original frame and the initial prediction generated by the first network, which acts as an attention mechanism enabling the second network to focus on interesting areas within the image, thereby improving the quality of its predictions. Quantitative evaluation carried out on several polyp segmentation databases shows that double encoder-decoder networks clearly outperform their single encoder-decoder counterparts in all cases. In addition, our best double encoder-decoder combination attains excellent segmentation accuracy and reaches state-of-the-art performance results in all the considered datasets, with a remarkable boost of accuracy on images extracted from datasets not used for training.

show abstract

Polyp Segmentation with Fully Convolutional Deep Neural Networks—Extended Evaluation Study

Cited by 41 publications

References 61 publications

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation

Colorectal multi-class image classification using deep learning models

Double Encoder-Decoder Networks for Gastrointestinal Polyp Segmentation

Contact Info

Product

Resources

About