Comparative diagnostic accuracy of InfraScanner in paediatric versus adult traumatic brain injury: A prospective, computed tomography–validated study

Objective: To compare the diagnostic accuracy of InfraScanner against computed tomography (CT) for detection of intracranial haematoma in paediatric versus adult patients with traumatic brain injury (TBI), and to identify lesion-level predictors of device detection. Methodology: This prospective observational, CT-validated study recruited 400 consecutive TBI patients (200 paediatric, 200 adult) at the Department of Neurosurgery, Lady Reading Hospital – Medical Teaching Institution (MTI), Peshawar from January 2025 to June 2025. Blinded InfraScanner Model 2000 and CT assessments were performed within one hour of each other. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated using 95% Wilson confidence intervals (CIs). Proportions were compared with the chi-squared test. Predictors of InfraScanner detection among CT-positive cases were identified using multivariable logistic regression. Results: CT revealed a haematoma in 48 paediatric (24.00%) and 62 adult patients (31.00%). InfraScanner sensitivity was lower in children 62.50% (30/48; 95% CI 48.4–74.8) than in adults 80.65% (50/62; 95% CI 69.1–88.6; p = 0.034); specificity was higher in children 98.03% than in adults 89.86% (p = 0.003). Lesion volume greater than 3.5 mL (odds ratio [OR] 4.12; p < 0.001) and depth less than 2.5 cm (OR 3.56; p = 0.001) increased the odds of detection, whereas paediatric age reduced the odds of detection (OR 0.54; p = 0.031). Conclusion: InfraScanner shows age-differentiated diagnostic accuracy. Lower paediatric sensitivity limits stand-alone use in children. A positive result warrants immediate action. A negative result does not exclude pathology. InfraScanner is not a substitute for clinical assessment or CT.