INTRODUCTION: The evaluation of endoscopic disease severity is a crucial component in managing patients with ulcerative colitis (UC). However, endoscopic assessment suffers from substantial intraobserver and interobserver variations, limiting the reliability of individual assessments. Therefore, we aimed to develop a deep learning model capable of distinguishing active from healed mucosa and differentiating between different endoscopic disease severity degrees.
METHODS: One thousand four hundred eighty-four unique endoscopic images from 467 patients were extracted for classification. Two experts classified all images independently of one another according to the Mayo endoscopic subscore (MES). In cases of disagreement, a third expert classified the images. Different convolutional neural networks were considered for automatically classifying UC severity. Five-fold cross-validation was used to develop and select the final model. Afterward, unseen test data sets were used for model evaluation.
RESULTS: In the most challenging task-distinguishing between all categories of MES-our final model achieved a test accuracy of 0.84. When evaluating this model on the binary tasks of distinguishing MES 0 vs 1-3 and 0-1 vs 2-3, it achieved accuracies of 0.94 and 0.93 and areas under the receiver operating characteristic curves of 0.997 and 0.998, respectively.
DISCUSSION: We have developed a highly accurate, new, automated way of evaluating endoscopic images from patients with UC. We have demonstrated how our deep learning model is capable of distinguishing between all 4 MES levels of activity. This new automated approach may optimize and standardize the evaluation of disease severity measured by the MES across centers no matter the level of medical expertise.
- Colitis, Ulcerative/diagnostic imaging
- Intestinal Mucosa
- Neural Networks, Computer
- Reproducibility of Results
- Severity of Illness Index