Abstract
Purpose In order to resolve the situation of high missed diagnosis rate and high misdiagnosis rate of the pathological analysis of the gastrointestinal endoscopic images by experts, we propose an automatic polyp detection algorithm based on Single Shot Multibox Detector (SSD). Method In the paper, SSD is based on VGG-16, the fully connected layer is changed to a convolutional layer, and four convolutional layers with successively decreasing scales are added as a new network structure. In order to verify the practicability, it is not only compared with manual polyp detection but also with Mask R-CNN. Results Multiple experimental results show that the mean Average Precision (mAP) of the SSD network is 95.74%, which is 12.4% higher than the manual detection and 5.7% higher than the Mask R-CNN. When detecting a single frame of image, the detection speed of SSD is 8.41 times that of manual detection. Conclusion Based on the traditional pattern recognition algorithm and the target detection algorithm using deep learning, we select a variety of algorithms to identify and classify polyps to achieve efficient detection results. Our research demonstrates that deep learning has a lot of room for development in the field of gastrointestinal image recognition.
Highlights
Endoscope technology is widely used in the diagnosis of gastrointestinal diseases [1,2,3]
In 2014, Girshick et al proposed a deep learning target detection algorithm RCNN [9] based on a region of interest [10] combined with a convolutional neural network (CNN) [11,12,13], which made a breakthrough in target detection
In order to break the situation of high missed diagnosis rate and high misdiagnosis rate of the pathological part of the gastrointestinal endoscopy image recognized by experts with naked eyes, we propose an algorithm for automatic polyp detection based on Single Shot Multibox Detector (SSD) [28]
Summary
Endoscope technology is widely used in the diagnosis of gastrointestinal diseases [1,2,3]. A large number of medical images will be generated during the detection process It is a very time-consuming and laborious task to only rely on the doctor’s naked eyes to identify the lesion-containing part from the large number of gastrointestinal endoscopic images [4,5,6,7], and the diagnosis process mainly relies on the doctor’s experience and pathology. Each endoscopy will produce a large number of images, and most of them do not contain lesion information. Accurate detection of lesions in medical images provides a guarantee of diagnostic information for clinical applications. This topic is useful for assisting doctors in screening. With the rise of machine learning and artificial intelligence, computer vision has been further
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
