• Corpus ID: 239016961

Self-Supervised U-Net for Segmenting Flat and Sessile Polyps

  title={Self-Supervised U-Net for Segmenting Flat and Sessile Polyps},
  author={Debayan Bhattacharya and Christian Betz and Dennis Eggert and A. Schlaefer},
Colorectal Cancer(CRC) poses a great risk to public health. It is the third most common cause of cancer in the US. Development of colorectal polyps is one of the earliest signs of cancer. Early detection and resection of polyps can greatly increase survival rate to 90%. Manual inspection can cause misdetections because polyps vary in color, shape, size and appearance. To this end, Computer-Aided Diagnosis systems(CADx) has been proposed that detect polyps by processing the colonoscopic videos… 

Figures and Tables from this paper


A Comprehensive Study on Colorectal Polyp Segmentation With ResUNet++, Conditional Random Field and Test-Time Augmentation
Improvements to the overall prediction performance of the ResUNet++ architecture can be achieved by using Conditional Random Field (CRF) and Test-Time Augmentation (TTA) and it is demonstrated that applying CRF and TTA improves the performance on various polyp segmentation datasets both on the same dataset and cross-dataset.
A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Images
A comparative study is performed to show that FCNs significantly outperform, without any further postprocessing, prior results in endoluminal scene segmentation, especially with respect to polyp segmentation and localization.
ResUNet++: An Advanced Architecture for Medical Image Segmentation
ResUNet++ is proposed, which is an improved ResUNet architecture for colonoscopic image segmentation, which significantly outperforms U-Net and Res UNet, two key state-of-the-art deep learning architectures, by achieving high evaluation scores.
Kvasir-SEG: A Segmented Polyp Dataset
This paper presents Kvasir-SEG: an open-access dataset of gastrointestinal polyp images and corresponding segmentation masks, manually annotated by a medical doctor and then verified by an experienced gastroenterologist, and demonstrates the use of the dataset with a traditional segmentation approach and a modern deep-learning based Convolutional Neural Network approach.
The Miss Rate for Colorectal Adenoma Determined by Quality-Adjusted, Back-to-Back Colonoscopies
There was a significant miss rate in the detection of colonic adenomas even in quality-adjusted, back-to-back colonoscopies, suggesting that the adenoma miss rate can be reduced with a sufficient observation time during colonoscopic insertion.
Attention U-Net: Learning Where to Look for the Pancreas
A novel attention gate (AG) model for medical imaging that automatically learns to focus on target structures of varying shapes and sizes is proposed to eliminate the necessity of using explicit external tissue/organ localisation modules of cascaded convolutional neural networks (CNNs).
Right-Sided Location Not Associated with Missed Colorectal Adenomas in an Individual-Level Reanalysis of Tandem Colonoscopy Studies.
It is found that right-side location of an adenomas does not increase its odds for being missed during colonoscopy, but adenoma size and histologic features do increase risk.
A review of computer-aided diagnosis in thoracic and colonic imaging.
  • Kenji Suzuki
  • Medicine
    Quantitative imaging in medicine and surgery
  • 2012
In this review, CAD of the thorax and colon, including CAD for detection and diagnosis of lung nodules in thoracic CT, and that for detection of polyps in CT colonography, are reviewed.
Tversky Loss Function for Image Segmentation Using 3D Fully Convolutional Deep Networks
A generalized loss function based on the Tversky index is proposed to address the issue of data imbalance and achieve much better trade-off between precision and recall in training 3D fully convolutional deep neural networks.
Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation
A Recurrent Convolutional Neural Network (RCNN) based on U-Net as well as a Recurrent Residual convolutional neural Network (RRCNN), which are named RU-Net and R2U-Net respectively are proposed, which show superior performance on segmentation tasks compared to equivalent models including U-nets and residual U- net.