scholarly journals Iterative deep learning for improved segmentation of endoscopic images

2021 ◽  
Vol 1 (1) ◽  
pp. 38-40
Author(s):  
Sharib Ali ◽  
Nikhil K Tomar
Keyword(s):  
Deep Learning ◽  
Iterative Procedure ◽  
Endoscopic Imaging ◽  
Endoscopic Images ◽  
Segmentation Strategy ◽  
Inference Techniques

Iterative segmentation is a unique way to prune the segmentation maps initialized by faster inference techniques or even unsupervised traditional thresholding methods. We used our previous feedback attention-based method for this work and demonstrate that with optimal iterative procedure our method can reach competitive accuracies in endoscopic imaging. For this work, we have applied this segmentation strategy for polyps and instruments.

Endoscopic prediction of submucosal invasion in Barrett’s cancer with the use of Artificial Intelligence: A pilot Study

Endoscopy ◽  
2020 ◽  
Author(s):  
Alanna Ebigbo ◽  
Robert Mendel ◽  
Tobias Rückert ◽  
Laurin Schuster ◽  
Andreas Probst ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Deep Learning ◽  
Pilot Study ◽  
White Light ◽  
Submucosal Invasion ◽  
Learning System ◽  
Endoscopic Images ◽  
Barrett’S Cancer ◽  
Significant Difference ◽  
Sensitivity Specificity

Background and aims: The accurate differentiation between T1a and T1b Barrett’s cancer has both therapeutic and prognostic implications but is challenging even for experienced physicians. We trained an Artificial Intelligence (AI) system on the basis of deep artificial neural networks (deep learning) to differentiate between T1a and T1b Barrett’s cancer white-light images. Methods: Endoscopic images from three tertiary care centres in Germany were collected retrospectively. A deep learning system was trained and tested using the principles of cross-validation. A total of 230 white-light endoscopic images (108 T1a and 122 T1b) was evaluated with the AI-system. For comparison, the images were also classified by experts specialized in endoscopic diagnosis and treatment of Barrett’s cancer. Results: The sensitivity, specificity, F1 and accuracy of the AI-system in the differentiation between T1a and T1b cancer lesions was 0.77, 0.64, 0.73 and 0.71, respectively. There was no statistically significant difference between the performance of the AI-system and that of human experts with sensitivity, specificity, F1 and accuracy of 0.63, 0.78, 0.67 and 0.70 respectively. Conclusion: This pilot study demonstrates the first multicenter application of an AI-based system in the prediction of submucosal invasion in endoscopic images of Barrett’s cancer. AI scored equal to international experts in the field, but more work is necessary to improve the system and apply it to video sequences and in a real-life setting. Nevertheless, the correct prediction of submucosal invasion in Barret´s cancer remains challenging for both experts and AI.

scholarly journals Multi-Task Model for Esophageal Lesion Analysis Using Endoscopic Images: Classification with Image Retrieval and Segmentation with Attention

Sensors ◽  
2021 ◽  
Vol 22 (1) ◽  
pp. 283
Author(s):  
Xiaoyuan Yu ◽  
Suigu Tang ◽  
Chak Fong Cheang ◽  
Hon Ho Yu ◽  
I Cheong Choi
Keyword(s):  
Deep Learning ◽  
Image Retrieval ◽  
Learning Model ◽  
Endoscopic Images ◽  
Proposed Model ◽  
Esophageal Lesion ◽  
Lesion Analysis ◽  
Diagnosis Accuracy ◽  
Segmentation Task ◽  
Deep Learning Model

The automatic analysis of endoscopic images to assist endoscopists in accurately identifying the types and locations of esophageal lesions remains a challenge. In this paper, we propose a novel multi-task deep learning model for automatic diagnosis, which does not simply replace the role of endoscopists in decision making, because endoscopists are expected to correct the false results predicted by the diagnosis system if more supporting information is provided. In order to help endoscopists improve the diagnosis accuracy in identifying the types of lesions, an image retrieval module is added in the classification task to provide an additional confidence level of the predicted types of esophageal lesions. In addition, a mutual attention module is added in the segmentation task to improve its performance in determining the locations of esophageal lesions. The proposed model is evaluated and compared with other deep learning models using a dataset of 1003 endoscopic images, including 290 esophageal cancer, 473 esophagitis, and 240 normal. The experimental results show the promising performance of our model with a high accuracy of 96.76% for the classification and a Dice coefficient of 82.47% for the segmentation. Consequently, the proposed multi-task deep learning model can be an effective tool to help endoscopists in judging esophageal lesions.

scholarly journals Optical endovascular imaging combining endoscopy, NBI and OCT, a feasibility study

2019 ◽  
Vol 5 (1) ◽  
pp. 577-580
Author(s):  
Axel Boese ◽  
Akhil Karthasseril Sivankutty ◽  
Michael Friebe
Keyword(s):  
Blood Flow ◽  
Structural Information ◽  
Endoscopic Imaging ◽  
Camera System ◽  
Video Mode ◽  
Endoscopic Images ◽  
System A ◽  
Analysis Of The Images ◽  
Definition Of ◽  
Short Time

AbstractFor imaging of the vascular structure, angiography is state of the art. This can be done by contrast enhanced XRay, CT or MR imaging. But these modalities typically only show the blood flow and do not allow a depiction of the vasculature itself. To provide information about the vessel walls and plaques narrowing the blood flow, catheter based intra vascular ultrasound or vascular optical coherence tomography can be used. Optical endoscopic imaging is rarely used in vascular diagnosis. But endoscopic imaging can depict superficial inflammations or defects of the intima vessel layer and the real anatomical shape of the inner vasculature e.g. at bifurcations or aneurysms. Since OCT and endoscopic imaging both need a flushing to remove the blood for a short time, a combination of both modalities seems viable. For combining the two modalities, various background studies were performed including the selection of a feasible fibre endoscope, light source and camera system. A new pull-back and flushing device was designed and created for realizing the synchronous image acquisition using the two modalities. For calibration of the system and definition of the pullback and imaging parameters, first tests on artificial phantoms were performed. Then vascular and tissue models were imaged in a combined pullback mode after using the flush for complete blood removal. Endoscopic images were acquired in a video mode. The analysis of the images was done subjectively. As expected, the OCT provided structural information of the wall. The endoscopic images in combination with pullback appear blurry in video mode. The flushing liquid hinders the automatic focusing of the camera. Thus, smaller details could not be identified but bifurcations were visible. Even though the results were not good as expected, the study showed the potential of a bimodal system and addressed the issues faced in the initial implementation.

Diagnosis of ulcerative colitis from endoscopic images based on deep learning

2022 ◽  
Vol 73 ◽  
pp. 103443
Author(s):  
Xudong Luo ◽  
Junhua Zhang ◽  
Zonggui Li ◽  
Ruiqi Yang
Keyword(s):  
Ulcerative Colitis ◽  
Deep Learning ◽  
Endoscopic Images

Spotting malignancies from gastric endoscopic images using deep learning

2019 ◽  
Vol 33 (11) ◽  
pp. 3790-3797 ◽  
Author(s):  
Jang Hyung Lee ◽  
Young Jae Kim ◽  
Yoon Woo Kim ◽  
Sungjin Park ◽  
Youn-i Choi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Endoscopic Images

scholarly journals DETECTION AND SEGMENTATION OF ENDOSCOPIC ARTEFACTS AND DISEASES USING DEEP ARCHITECTURES

2020 ◽  
Author(s):  
Nhan T. Nguyen ◽  
Dat Q. Tran ◽  
Dung B. Nguyen
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
Detection Task ◽  
Mean Average Precision ◽  
Model Ensemble ◽  
Average Precision ◽  
Endoscopic Images ◽  
Feature Pyramid ◽  
Segmentation Task ◽  
Multiple Detectors

ABSTRACTWe describe in this paper our deep learning-based approach for the EndoCV2020 challenge, which aims to detect and segment either artefacts or diseases in endoscopic images. For the detection task, we propose to train and optimize EfficientDet—a state-of-the-art detector—with different EfficientNet backbones using Focal loss. By ensembling multiple detectors, we obtain a mean average precision (mAP) of 0.2524 on EDD2020 and 0.2202 on EAD2020. For the segmentation task, two different architectures are proposed: UNet with EfficientNet-B3 encoder and Feature Pyramid Network (FPN) with dilated ResNet-50 encoder. Each of them is trained with an auxiliary classification branch. Our model ensemble reports an sscore of 0.5972 on EAD2020 and 0.701 on EDD2020, which were among the top submitters of both challenges.

Sign in / Sign up

Export Citation Format

Share Document