Mathematical Modelling of Orofacial Infection Spread Dynamics to Predict Fascial Space Involvement in Odontogenic Infections in Nigeria

Odontogenic infections can spread along anatomically predictable pathways through interconnected fascial spaces of the head and neck, with potentially life-threatening consequences when multiple spaces are involved. While clinical experience guides most management decisions, no predictive mathematical model exists for anticipating the pattern and rate of fascial space involvement based on initial clinical parameters in the sub-Saharan African setting. This PhD study develops and validates a deterministic mathematical model of orofacial infection spread dynamics calibrated to clinical data from Nigerian tertiary institutions. Using retrospective clinical data from 340 odontogenic infection cases managed at six teaching hospitals across Nigeria's six geopolitical zones between 2018 and 2023, differential equation-based compartmental models were fitted to describe transition probabilities between anatomical compartments (primary space, secondary space, mediastinal extension) as functions of infecting organism virulence proxy, host immune status (assessed by serum albumin, lymphocyte count, HbA1c), and time from symptom onset. Model parameters were estimated using Bayesian inference, and the fitted model was prospectively validated against 60 new cases. The model achieved 79.3% accuracy in predicting secondary space involvement within 48 hours of initial assessment. Diabetic patients and those presenting beyond 72 hours from symptom onset showed accelerated infection spread trajectories in the model. The study makes a novel interdisciplinary contribution to oral surgical decision science and recommends the development of a clinical decision support tool based on the model for use in emergency oral surgery settings across Nigeria. Keywords: mathematical modelling, odontogenic infections, fascial space, spread dynamics, Nigeria.