Post written by Wanqing Xie, PhD, from the Department of Intelligent Medical Engineering, School of Biomedical Engineering, Anhui Medical University, Hefei, China, and Beth Israel Deaconess Medical Center, Harvard Medical School, Harvard University, Boston, Massachusetts, USA.

We studied the lesion classification of small-bowel Crohn’s disease (CD) and ulcer severity grading in terms of depth, size, and area by using artificial intelligence.

Accurate diagnosis of small-bowel CD is very important for CD patients. However, in practice, because of problems such as shortage of medical resources and differences in the level of physicians, it is difficult to achieve a high diagnostic rate for small-bowel CD.

Therefore, we considered it meaningful and necessary to study an intelligent diagnosis system that can effectively improve the small-bowel CD diagnostic rate, especially in lesion detection and grading, for better CD management.

Based on the deep learning model (EfficientNet-b5), our study performed automated classification and grading of CD lesions on double-balloon endoscopy images with high accuracy and robustness.

In future work, we will further consider directly processing video double-balloon endoscopy data for small-bowel CD to better assist endoscopists in examination.

We will continue our study on intelligence-assisted diagnosis of small-bowel CD, and we sincerely hope to share more valuable research.

Grad-CAM visualization for different lesions. A, Ulcers. B, Noninflammatory stenosis. C, Inflammatory stenosis.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.