ORAHS2024
Abstract Submission

151. Longitudinal follow-up of patients with IBD, focusing on aggressive front-loading drug escalation

Contributed abstract in session TC-1: Poster session, stream Posters.

Tuesday, 14:00-15:30
Room: Auditorium

Authors (first author is the speaker)

1. Thomas Xenos
School of Production Engineering and Management, Technical University of Crete
2. Ioannis Drygiannakis
Medicine, University of Crete
3. Nikolaos Matsatsinis
Department of Production Engineering and Management, Technical University of Crete
4. Ioannis Koutroubakis
Gastroenterology Department, University Hospital of Heraklion · Heraklion, GRC

Abstract

Here, IBD patients are studied over time to analyse trade-off factors that determine disease burden, being classified on criteria related to its activity. Setting up a model to predict disease persistence and enhance decision making towards aggressive treatment with biologics is important. We aim to accelerate and maximise healing probability, deter disease relapse while reduce side effects avoiding surgery and other debilitating conditions. Similar issues were confronted with methods describing data in unambiguous way, comparing observed values to expected results· however, considering neither time variables nor multiparametric nature of predictive criteria incorporated. Clustering by disease phenotype into abnormal biomarkers is the gold standard for treatment decision making in intestinal inflammation. Hence, multiparametric approach entails specific predictive criteria with concurrent incorporation of time, reflecting choices. Thus, when diseases are modified, we face an urgent need to realign required treatment and intervene to adapt the new conditions. Robust multicriteria models support us in optimising qualitative, quantitative clinical and laboratory cut-off values at which treatment escalation is both essential and effective. Usage of Ensemble Methods from Machine Learning and Multi-Criteria Decision Analysis, helps in the classification, based on recurrence risk assessment criteria, achieving a consistent classification of patients in need of advanced treatments.

Keywords

Status: accepted

Back to the list of papers