In recent years, object detection technology in the field of artificial intelligence has made rapid advancements. Various industries have integrated their respective sectors with object detection technology, making them more intelligent and efficient. Due to its significant applications in computer-aided diagnosis and computer-aided detection techniques, an increasing number of researchers are transferring object detection technology to the medical field. Colonic endoscopy is an essential approach for detecting gastrointestinal diseases, however, the phenomenon of missed and false detections is still difficult to avoid when using white-light endoscopy. In order to improve the lesion detection rate for colorectal cancer patients, object detection technology is being applied to endoscopic examinations. And the YOLO series models have been the preferred choice for computer-aided diagnosis due to their high accuracy. However, the requirement of the NMS method for post-processing redundant bounding boxes in YOLO models leads to training and inference delays, which cannot meet the real-time demands in clinical practice. This paper proposes a real-time lesion diagnosis model, RTDETRC, based on RT-DETR. In this diagnostic model, colonic lesions are classified into three categories: polyps, adenomas, and malignant tumors. The Mixup strategy is employed during training to enhance the model's robustness. Additionally, channel attention mechanism and spatial attention mechanism are introduced in the neck part of the model to enhance information interaction among features. With a lower parameter cost, learn richer feature vectors, and improve both the model's inference speed and detection performance. The experimental results show that the proposed lesion diagnosis model has a faster and more accurate lesion detection capability.And the model improves mAP@50 and mAP@50-95 to 0.880 and 0.721, respectively, The inference time was reduced to 5.9ms.
As a device for detecting colopathy, white light endoscopy still suffers from missed diagnoses and misdiagnoses. For increasing the diagnostic rate of colopathy sick persons, our work proposes a new diagnoses model for pathological changes based on an improved YOLOv5 algorithm based on the SwinTransformer framework. In this diagnostic model, colopathy is divided into three categories: polyp of the colon, adenoma of the colon, and carcinomas. The SwinStage structure, as well as the channel attention mechanism and spatial attention mechanism, are incorporated to fully extract features of the map and the relationship between feature maps, thereby improving the result of object detection. Experimental outcomes demonstrate that the proposed pathological tissue diagnoses model has a more excellent pathological tissue detection ability, the mAP@0.5 and mAP@0.5:0.95 can attain 0.915 and 0.717.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.