177. Assessing Regional Sagittal Craniosynostosis Severity Using the CraniorateTM Machine Learning Toolset

PURPOSE: Sagittal craniosynostosis (SC) can affect three distinct regions of the skull: frontal, central, and posterior. Existing literature mainly focuses on manually defined anatomical measurements to quantify these different components. CranioRateTM is an advanced supervised machine learning toolset able to characterize and quantify craniosynostosis severity, validated by expert surgeon ratings. The aim of this study is to describe the development of regional severity scores (rSSS) using the CranioRate machine learning toolset. METHODS: Our study population included patients with and without SC. Demographic information was collected. The sagittal severity scores were predicted using a regression model, mapping skull shape descriptors to 50 consensus expert ratings. Principal component regression was used to generate a model of regional skull variation in a large cohort of patients with SC and a matched control population. These regions were divided and decomposed to assess the regional component of sagittal severity. Receiver Operating Characteristics (ROC) were conducted to assess the AUC (Area Under curve) for rSSS. RESULTS: The experimental group had 195 patients with SC, and the control group had 170 without SC. Demographic variables were similar in both groups. Regression pattens formed three distinct regions: frontal, central, and posterior encompassing 94.3% of skull morphology. Another baseline region was formed from the remaining morphology. The AUC for comparing the 3 rSSS to control skulls was 0.997 for frontal, 0.998 for central, and 0.994 for posterior regions. CONCLUSION: The very high AUC values for the rSSS indicate that the regional scores are highly effective at identifying the presence of sagittal craniosynostosis (SC). CranioRate’s advanced machine learning algorithm allows for precise and comprehensive measurement of regional sagittal severity. This tool can help surgeons educate patients, conduct research, and enhance clinical care, adding regional insight to the objective sagittal severity score.