Risk prediction model based on blood biomarkers for predicting moderate to severe endoscopic activity in patients with ulcerative colitis

Utility of four machine learning approaches for identifying ulcerative colitis and Crohn's disease

Objective

Peripheral blood routine parameters (PBRPs) are simple and easily acquired markers to identify ulcerative colitis (UC) and Crohn's disease (CD) and reveal the severity, whereas the diagnostic performance of individual PBRP is limited. We, therefore used four machine learning (ML) models to evaluate the diagnostic and predictive values of PBRPs for UC and CD.

Methods

A retrospective study was conducted by collecting the PBRPs of 414 inflammatory bowel disease (IBD) patients, 423 healthy controls (HCs), and 344 non-IBD intestinal diseases (non-IBD) patients. We used approximately 70 % of the PBRPs data from both patients and HCs for training, 30 % for testing, and another group for external verification. The area under the receiver operating characteristic curve (AUC) was used to evaluate the diagnosis and prediction performance of these four ML models.

Results

Multi-layer perceptron artificial neural network model (MLP-ANN) yielded the highest diagnostic performance than the other three models in six subgroups in the training set, which is helpful for discriminating IBD and HCs, UC and CD, active CD and remissive CD, active UC and remissive UC, non-IBD and HCs, and IBD and non-IBD with the AUC of 1.00, 0.988, 0.942, 1.00, 0.986, and 0.97 in the testing set, as well as the AUC of 1.00, 1.00, 0.773, 0.904, 1.00 and 0.992 in the external validation set.

Conclusion

PBRPs-based MLP-ANN model exhibited good performance in discriminating between UC and CD and revealing the disease activity; however, a larger sample size and more models need to be considered for further research.

Clinical value of the Toronto inflammatory bowel disease global endoscopic reporting score in ulcerative colitis

BACKGROUND

Endoscopic evaluation in diagnosing and managing ulcerative colitis (UC) is becoming increasingly important. Several endoscopic scoring systems have been established, including the Ulcerative Colitis Endoscopic Index of Severity (UCEIS) score and Mayo Endoscopic Subscore (MES). Furthermore, the Toronto Inflammatory Bowel Disease Global Endoscopic Reporting (TIGER) score for UC has recently been proposed; however, its clinical value remains unclear.

AIM

To investigate the clinical value of the TIGER score in UC by comparing it with the UCEIS score and MES.

METHODS

This retrospective study included 166 patients with UC who underwent total colonoscopy between January 2017 and March 2023 at the Affiliated Hospital of Qingdao University (Qingdao, China). We retrospectively analysed endoscopic scores, laboratory and clinical data, treatment, and readmissions within 1 year. Spearman's rank correlation coefficient, receiver operating characteristic curve, and univariate and multivariable logistic regression analyses were performed using IBM SPSS Statistics for Windows, version 26.0 (IBM Corp., Armonk, NY, United States) and GraphPad Prism version 9.0.0 for Windows (GraphPad Software, Boston, Massachusetts, United States).

RESULTS

The TIGER score significantly correlated with the UCEIS score and MES (r = 0.721, 0.626, both P < 0.001), showed good differentiating values for clinical severity among mild, moderate, and severe UC [8 (4-112.75) vs 210 (109-219) vs 328 (219-426), all P < 0.001], and exhibited predictive value in diagnosing patients with severe UC [area under the curve (AUC) = 0.897, P < 0.001]. Additionally, the TIGER (r = 0.639, 0,551, 0.488, 0.376, all P < 0.001) and UCEIS scores (r = 0.622, 0,540, 0.494, and 0.375, all P < 0.001) showed stronger correlations with laboratory and clinical parameters, including C-reactive protein, erythrocyte sedimentation rate, length of hospitalisation, and hospitalisation costs, than MES (r = 0.509, 0,351, 0.339, and 0.270, all P < 0.001). The TIGER score showed the best predictability for patients' recent advanced treatment, including systemic corticosteroids, biologics, or immunomodulators (AUC = 0.848, P < 0.001) and 1-year readmission (AUC = 0.700, P < 0.001) compared with the UCEIS score (AUC = 0.762, P < 0.001; 0.627, P < 0.05) and MES (AUC = 0.684, P < 0.001; 0.578, P = 0.132). Furthermore, a TIGER score of ≥ 317 was identified as an independent risk factor for advanced UC treatment (P = 0.011).

CONCLUSION

The TIGER score may be superior to the UCIES score and MES in improving the accuracy of clinical disease severity assessment, guiding therapeutic decision-making, and predicting short-term prognosis.