| Autor: |
Byeon, J. S., Park, K., Lim, J., Kim, N. |
| Předmět: |
|
| Zdroj: |
Endoscopy; 2024 Supplement 2, Vol. 56, pS113-S113, 1p |
| Abstrakt: |
This article discusses a study that aimed to develop an artificial intelligence algorithm to assist in the differential diagnosis between Crohn's disease (CD) and intestinal tuberculosis (ITB) using colonoscopy images. The study used a convolutional neural network (CNN) model and a dataset of colonoscopy images from patients diagnosed with CD or ITB. The developed model exhibited high accuracy and area under the receiver operating characteristic curve (AUROC) in both the test dataset and an external dataset. The CNN model performed slightly better than trainee endoscopists but was inferior to expert endoscopists. The study concludes that the CNN-based model has the potential to aid in the differential diagnosis between CD and ITB, particularly for less experienced endoscopists. [Extracted from the article] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
