Craniometrics in Metopic Craniosynostosis: A Review of Craniometric Parameters and the Emergence of Machine Learning Models.
Authors
Affiliations (5)
Affiliations (5)
- University of Colorado School of Medicine.
- Department of Surgery, University of Colorado Anschutz Medical Campus.
- Department of Surgery, Division of Plastic and Reconstructive Surgery, University of Colorado Anschutz Medical Campus.
- Department of Biostatistics and Informatics, Colorado School of Public Health, University of Colorado Anschutz Medical Campus.
- Department of Pediatric Plastic and Reconstructive Surgery, Children's Hospital Colorado, University of Colorado Anschutz Medical Campus, Aurora, CO.
Abstract
Metopic craniosynostosis (MCS) is defined by premature fusion of the metopic suture, resulting in trigonocephaly and potential neurodevelopmental implications. Advances in diagnosis and treatment over the past 2 decades have improved landmarking and evaluation using objective imaging metrics. While computed tomography (CT) remains the diagnostic standard, concerns over pediatric radiation exposure have increased reliance on 3D photogrammetry. Both modalities, however, have limitations in capturing the full range of craniofacial dysmorphology. This systematic review evaluates current craniometric parameters and emerging machine learning (ML) models used to assess MCS morphology. A systematic search of PubMed and Google Scholar, following PRISMA guidelines, identified English-language studies reporting imaging-based craniometric or ML methods for MCS assessment. Extracted outcomes included severity metrics, roles in surgical decision-making, and postoperative evaluation. Due to study heterogeneity, findings were descriptively synthesized. Fifty-eight studies involving 9068 patients met the inclusion criteria. A total of 2425 (26.7%) had MCS. CT was the most common imaging modality (78.4%), followed by 3D (15.7%) and 2D (7.8%) photogrammetry. Over 100 craniometric parameters were reported, most frequently the interfrontal angle (IFA) and endocranial bifrontal angle (EBA). Eighteen studies (31%) utilized ML models introducing indices such as the Metopic Severity Score, Cranial Morphology Deviation score, and Head Shape Anomaly index, which showed high diagnostic accuracy for severity grading and outcome prediction. CT-based metrics remain standard, but ML models using advanced imaging offer radiation-free, objective, and reproducible assessments. Future efforts should emphasize multicenter data sharing, standardized variables, longitudinal imaging, and integration of genotypic and neuropsychological data.