Related Experiment Video
Updated: May 21, 2025
15:49
Flexible Colonoscopy in Mice to Evaluate the Severity of Colitis and Colorectal Tumors Using a Validated Endoscopic Scoring System
Published on: October 16, 2013
31.7K
Cascade-E-Yolov5s network for recognizing the ulcerative lesion subtypes in small intestinal.
Xudong Guo1, Liying Pang1, Lei Xu1
1School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.
The Review of Scientific Instruments
|March 18, 2025
Summary
Accurately diagnosing small intestinal ulcers is challenging. A new AI model, Cascade-E-Yolov5s, significantly improves immediate detection and classification of ulcer subtypes, aiding clinical diagnosis.
Area of Science:
- Gastroenterology and Artificial Intelligence
- Medical Imaging and Diagnostics
- Computational Pathology
Background:
- Accurate diagnosis of small intestinal ulcers is difficult due to lesion complexity and distribution, often causing diagnostic delays.
- Current diagnostic methods rely on pathology and follow-up, limiting immediate diagnostic accuracy.
- There is a need for advanced tools to improve the efficiency and precision of early ulcer detection.
Purpose of the Study:
- To introduce and evaluate the Cascade-E-Yolov5s network for enhanced immediate diagnosis of small intestinal ulcer subtypes.
- To improve the accuracy and efficiency of identifying different types of small intestinal ulcers during endoscopic procedures.
- To provide a computational tool that assists clinicians in making faster and more precise diagnoses.
Main Methods:
- Development of the Cascade-E-Yolov5s network, integrating EfficientNet for image classification and SimAM-Yolov5s for lesion detection.
- Utilizing EfficientNet as a backbone within SimAM-Yolov5s, incorporating SIoU loss and a parameter-free attention module.
- Training and testing the model on a dataset of 4909 ulcer images from 684 patients, covering four distinct ulcer types.
Main Results:
- The Cascade-E-Yolov5s network achieved an average detection precision of 86.46%.
- The model obtained a mean average precision at an IoU threshold of 0.5 (mAP@0.5) of 82.20%.
- Cascade-E-Yolov5s demonstrated superior performance compared to conventional detection networks in identifying ulcer subtypes.
Conclusions:
- The Cascade-E-Yolov5s network effectively enhances the detection efficiency and accuracy of small intestinal ulcer subtypes.
- This AI model shows promise in assisting clinicians with more precise and timely diagnoses during endoscopy.
- The study highlights the potential of advanced deep learning models in overcoming diagnostic challenges in gastroenterology.
Related Concept Videos
01:27Histology of the Small Intestine
378
The small intestine exhibits a unique histological structure that significantly enhances its function in digestion and nutrient absorption. These structures include circular folds, villi, and various specialized cells that collectively facilitate the digestion of food.
The intestinal lining features transverse folds called circular folds, each housing fingerlike projections known as intestinal villi. These villi are covered by a layer of simple columnar epithelium, also referred to as...
The intestinal lining features transverse folds called circular folds, each housing fingerlike projections known as intestinal villi. These villi are covered by a layer of simple columnar epithelium, also referred to as...
378
01:23Renewal of Intestinal Stem Cells
2.5K
The intestinal epithelial lining rapidly renews every 4 to 5 days. The renewal is facilitated by intestinal stem cells (ISCs) located at the base of the crypt– a gland located at the bottom of each villus. ISCs divide asymmetrically to form new stem cells and progenitor daughter cells. The daughter cells are called transit-amplifying (TA) cells which move upwards along the crypt and either differentiate into absorptive cells– the enterocytes or secretory cells– including the...
2.5K