IDDF2018-ABS-0259 Segmentation of intestinal polyps via a deep learning algorithm

Liansheng Wang, Yuqi Qian, Yanxing HuClinical 2018 · DOI

Abstract

Colorectal cancer is a major global health concern, with the majority of cases originating from intestinal polyps. Early screening via Wireless Capsule Endoscopy (WCE) is vital but labor-intensive. This study proposes a Deep Convolutional Neural Network (DCNN) for automated polyp segmentation. Using the CVC-ColonDB and CVC-ClinicDB datasets, the model achieved an Intersection over Union (IoU) of over 0.5 on test data. The results demonstrate the potential of DCNNs to assist clinicians by providing automated visualization of polyps without manual feature extraction.

Introduction

Colorectal cancer ranks as the third most common cancer in men and the second in women. Since 80%–95% of these cancers evolve from intestinal polyps, early detection is the most effective prevention strategy. Wireless Capsule Endoscopy (WCE) provides a non-invasive means to examine the digestive tract; however, the sheer volume of images generated per patient creates a significant diagnostic burden. Computer-assisted segmentation offers a solution to help doctors visualize and identify polyps more efficiently.

Methods

The researchers implemented a Deep Convolutional Neural Network (DCNN) for the segmentation task. DCNNs are particularly effective for image segmentation as they learn hierarchical features directly from raw data, eliminating the need for manual preprocessing or "hand-crafted" feature extraction.

Data and Training

The model was trained and evaluated using two open-access datasets: CVC-ColonDB and CVC-ClinicDB. The data was split into training, validation, and test sets.

  • Loss Function: Cross-entropy loss was used for optimization.
  • Evaluation Metric: Performance was measured using the Intersection over Union (IoU) value, which quantifies the overlap between predicted masks and ground truth annotations.

Results

The quantitative evaluation shows that the model performs well across all subsets, though a performance gap exists between training and test data, likely due to limited data and high inter-patient variance.

Dataset Background IoU Polyp IoU Accuracy
Training 0.976 0.800 0.987
Validation 0.969 0.604 0.970
Test 0.947 0.523 0.725

As shown in the visual results in Figure 2, the model successfully identifies the polyp regions across various frames, producing segmentation masks that align closely with the ground truth.

Visual examples of polyp segmentation across different cases, showing original images, ground truth, and model predictions.

Conclusion

The experimental results confirm that the DCNN model is capable of effective polyp segmentation, achieving an IoU greater than 0.5 on unseen test data. A key advantage of this approach is its adaptability to polyps of different sizes and morphologies without requiring manual feature engineering.

Detailed segmentation results for three specific cases (a, b, c), demonstrating the model's robustness to varying polyp appearances and sizes.

Future work will focus on expanding the training datasets to improve generalization and exploring more advanced architectures to refine segmentation boundaries and improve performance on diverse clinical data.

Summary updated

Abstract

<h3>Background</h3> Colorectal is the third most commonly occurring cancer in men and the second most commonly occurring cancer in women. At least 80%–95% of colorectal cancers are evolved from intestinal polyps. Therefore early screening of intestinal polyps is one of the crucial ways to prevent colorectal cancer. WCE is widely used in the examination of the internal environment of the digestive tract. However, lots of medical images per patient undoubtedly create a considerable burden on doctors. Thus, in this paper, we describe a computer-assisted segmentation of intestinal polyps to help doctors visualise polyps in images. <h3>Methods</h3> We choose DCNN (Deep Convolution Neural Network) as our model, which has shown outstanding performance in image segmentation tasks in recent years. The data we used to train the model including two open intestinal polyp datasets, CVC-ColonDB and CVC-ClinicDB. The data are divided into training data, validation data and test data. At the same time, we use cross-entropy as a model-optimised loss function and the IoU value as our model’s evaluation criteria. <h3>Results</h3> Table 1 shows that our model performed well on these three data. At the same time, it should be noted that our model performed significantly better on the training data than the test data as shown in IDDF2018-ABS-0259 Figure 1. This may be due to the lack of data and the variance between different patients. <h3>Conclusions</h3> The experimental results show that our model has achieved good performance in the polyp segmentation task, which appears as an IoU value of more than 0.5 on the test set and visual examples illustrated in Figure 1. In addition, without the need for manual feature extraction and preprocessing, DCNN model shows good adaptation for polyps of different sizes.
BibTeX
@inproceedings{wang2018iddf,
  author = {Liansheng Wang and Yuqi Qian and Yanxing Hu},
  title = {{IDDF2018-ABS-0259 Segmentation of intestinal polyps via a deep learning algorithm}},
  booktitle = {Clinical},
  year = {2018},
  doi = {10.1136/gutjnl-2018-iddfabstracts.180},
}
Figure