Developing a Machine-Learning Prediction Model for Infliximab Response in Crohn’s Disease: Integrating Clinical Characteristics and Longitudinal Laboratory Trends
Bibliographic record
Abstract
BACKGROUND: Achieving long-term clinical remission in Crohn's disease (CD) with antitumor necrosis factor α (anti-TNF-α) agents remains challenging. AIMS: This study aims to establish a prediction model based on patients' clinical characteristics using a machine-learning approach to predict the long-term efficacy of infliximab (IFX). METHODS: Three cohorts comprising 746 patients with CD were included from 3 inflammatory bowel disease (IBD) centers between June 2013 and January 2022. Clinical records were collected from baseline, 14-, 30-, and 52-week post-IFX treatment. Three machine-learning approaches were employed to develop predictive models based on 23 baseline predictors. The SHapley Additive exPlanations (SHAP) algorithm was used to dissect underlying predictors, and latent class mixed model (LCMM) was applied for trajectory analysis of the longitudinal change of blood routine tests along with long-term IFX therapy. RESULTS: The XGBoost model exhibited the best discrimination between long-term responders and nonresponders. In the internal training and testing set, the model achieved an AUC of 0.91 (95% CI, 0.86-0.95) and 0.71 (95% CI, 0.66-0.87), respectively. Moreover, it achieved a moderate predictive performance in the independent external cohort, with an AUC of 0.68 (95% CI, 0.59-0.77). The SHAP algorithm revealed disease-relevant laboratory measurements, notably hemoglobin (HB), white blood cells (WBC), erythrocyte sedimentation rate (ESR), albumin (ALB), and platelets (PLT), alongside age at diagnosis and the Montreal classification, as the most influential predictors. Furthermore, 2 distinct patient clusters based on dynamic laboratory tests were identified for monitoring the long-term remission. CONCLUSIONS: The established prediction model demonstrated remarkable discriminatory power in distinguishing long-term responders from nonresponders to IFX therapy. The identification of distinct patient clusters further emphasizes the need for tailored therapeutic approaches in CD management.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".