To develop and validate machine learning (ML) models for objective and comprehensive quantification of sagittal craniosynostosis (SCS) severity, enhancing clinical assessment, management, and research.
Design
A cross-sectional study that combined the analysis of computed tomography (CT) scans and expert ratings.
Setting
The study was conducted at a children's hospital and a major computer imaging institution. Our survey collected expert ratings from participating surgeons.
Participants
The study included 195 patients with nonsyndromic SCS, 221 patients with nonsyndromic metopic craniosynostosis (CS), and 178 age-matched controls. Fifty-four craniofacial surgeons participated in rating 20 patients head CT scans.
Interventions
Computed tomography scans for cranial morphology assessment and a radiographic diagnosis of nonsyndromic SCS.
Main Outcomes
Accuracy of the proposed Sagittal Severity Score (SSS) in predicting expert ratings compared to cephalic index (CI). Secondary outcomes compared Likert ratings with SCS status, the predictive power of skull-based versus skin-based landmarks, and assessments of an unsupervised ML model, the Cranial Morphology Deviation (CMD), as an alternative without ratings.
Results
The SSS achieved significantly higher accuracy in predicting expert responses than CI (P < .05). Likert ratings outperformed SCS status in supervising ML models to quantify within-group variations. Skin-based landmarks demonstrated equivalent predictive power as skull landmarks (P < .05, threshold 0.02). The CMD demonstrated a strong correlation with the SSS (Pearson coefficient: 0.92, Spearman coefficient: 0.90, P < .01).
Conclusions
The SSS and CMD can provide accurate, consistent, and comprehensive quantification of SCS severity. Implementing these data-driven ML models can significantly advance CS care through standardized assessments, enhanced precision, and informed surgical planning.
KajdicNSpazzapanPVelnarT. Craniosynostosis—recognition, clinical characteristics, and treatment. Bosn J Basic Med Sci. 2018;18(2):110-116. doi:https://doi.org/10.17305/bjbms.2017.2083
3.
SpeltzMLCollettBRWallaceERStarrJRCradockMMBuonoLCunninghamMKapp-SimonK.Intellectual and academic functioning of school-age children with single-suture craniosynostosis. Pediatrics. 2015;135(3):e615-e623. doi:https://doi.org/10.1542/peds.2014-1634
4.
MassimiLCaldarelliMTamburriniGPaternosterGDi RoccoC. Isolated sagittal craniosynostosis: definition, classification, and surgical indications. Childs Nerv Syst.2012;28(9):1311-1317. doi:https://doi.org/10.1007/s00381-012-1834-5
5.
WesAMNaranSSunJMazzaferroDXuWNguyenPBartlettSPTaylorJA. The Whitaker classification of craniosynostosis outcomes: an assessment of interrater reliability. Plast Reconstr Surg.2017;140(4):579e-586e. doi:https://doi.org/10.1097/PRS.0000000000003688
6.
VillavisanisDFBlumJDChoDYBarreroCShakirSNahH-DSwansonJWTaylorJABartlettSP. Degree of sagittal suture fusion, cephalic index, and head shape in nonsyndromic sagittal craniosynostosis. J Craniofac Surg. 2022;33(8):2388-2393. doi:https://doi.org/10.1097/SCS.0000000000008782
7.
FishDHintonCBarriosA. Revisiting the cephalic index: the origin, purpose, and current applicability—a narrative review. JPO J Prosthet Orthot. 2024;36(2):e35-e48.
8.
BinsGPZhouLZCullDLaytonRGDunsonBTKoganSDavidLRRunyanCM. Creation of the scaphocephalic index: measurement of global and regional severity in scaphocephaly. Plast Reconstr Surg. 2024;154(2):335e-347e. doi:https://doi.org/10.1097/PRS.0000000000010845
9.
BlumJDChoDYCheungLVillavisanisDFSwansonJWBartlettSPTaylorJA. Making the diagnosis in sagittal craniosynostosis—it’s height, not length, that matters. Childs Nerv Syst.2022;38(7):1331-1340. doi:https://doi.org/10.1007/s00381-022-05518-3
10.
AndersonMGJungbauerDLeclairNKAhnESStoltzPMartinJEHershDSBooklandMJ. Incorporation of a biparietal narrowing metric to improve the ability of machine learning models to detect sagittal craniosynostosis with 2D photographs. Neurosurg Focus.2023;54(6):E9. doi:https://doi.org/10.3171/2023.3.FOCUS2349
11.
KronigODKronigSAVan AdrichemLN. New method for quantification of severity of isolated scaphocephaly linked to intracranial volume. Childs Nerv Syst.2021;37(4):1175-1183. doi:https://doi.org/10.1007/s00381-020-04932-9
12.
QamarABangiSFBarveR. Artificial intelligence applications in diagnosing and managing non-syndromic craniosynostosis: a comprehensive review. Cureus. 2023;15(9):e45318. doi:https://doi.org/10.7759/cureus.45318
13.
LuoAGursesMEGeciciNNKozelGLuVMKomotarRJIvanME. Machine learning applications in craniosynostosis diagnosis and treatment prediction: a systematic review. Childs Nerv Syst.2024;40(8):2535-2544. doi:https://doi.org/10.1007/s00381-024-06409-5
14.
BhalodiaRDvoracekLAAyyashAMKavanLWhitakerRGoldsteinJA. Quantifying the severity of metopic craniosynostosis: a pilot study application of machine learning in craniofacial surgery. J Craniofac Surg. 2020;31(3):697-701. doi:https://doi.org/10.1097/SCS.0000000000006215
15.
AnstadtEETaoWGuoEDvoracekLBruceMKGrossePJWangLKavanLWhitakerRGoldsteinJA. Quantifying the severity of metopic craniosynostosis using unsupervised machine learning. Plast Reconstr Surg.2023;151(2):396-403. doi:https://doi.org/10.1097/PRS.0000000000009853
16.
BeirigerJWTaoWBruceMKAnstadtEChristensenCSmetonaJWhitakerRGoldsteinJA. Craniorate: an image-based, deep-phenotyping analysis toolset and online clinician interface for metopic craniosynostosis. Plast Reconstr Surg.2024;153(1):112e-119e. doi:https://doi.org/10.1097/PRS.0000000000010452
17.
BeirigerJWTaoWBruceMKAnstadtEChristensenCSmetonaJWhitakerRGoldsteinJA. Craniorate: An image-based, deep-phenotyping analysis toolset and online clinician interface for metopic craniosynostosis. Accessed October 22, 2024. https://www.craniorate.org/
MattesDHaynorDRVesselleHLewellenTKEubankW. PET-CT image registration in the chest using free-form deformations. IEEE Trans Med Imaging.2003;22(1):120-128. doi:https://doi.org/10.1109/TMI.2003.809072
20.
CatesJElhabianSWhitakerR. Shapeworks: particle-based shape correspondence and visualization software. Stat Shape Deform Anal. 2017:257-298.
21.
BruceMKTaoWBeirigerJChristensenCPfaffMJWhitakerRGoldsteinJA. 3D photography to quantify the severity of metopic craniosynostosis. Cleft Palate Craniofac J. 2023;60(8):971-979. doi:https://doi.org/10.1177/10556656221087071
RevelleWRevelleMW. Package ‘psych’. Comprehen R Arch Netw. 2015;337(338):161-165.
24.
ChalmersRP. Mirt: a multidimensional item response theory package for the R environment. J Stat Softw.2012;48(6):1-29. doi:https://doi.org/10.18637/jss.v048.i06
25.
TharwatAGaberTIbrahimAHassanienAE. Linear discriminant analysis: a detailed tutorial. AI Commun.2017;30(2):169-190.
26.
TeamRC. R: A language and environment for statistical computing, R Foundation for Statistical. Computing. 2020.
27.
LakensD. Equivalence tests: a practical primer for t tests, correlations, and meta-analyses. Soc Psychol Pers Sci.2017;8(4):355-362. doi:https://doi.org/10.1177/1948550617697177
28.
VirtanenPGommersROliphantTEHaberlandMReddyTCournapeauDBurovskiEPetersonPWeckesserWBrightJ, et al. Scipy 1.0: fundamental algorithms for scientific computing in python. Nat Methods.2020;17(3):261-272.
CicchettiDV. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess.1994;6(4):284-290.
31.
AlzahraniSAAlzahraniAAAl-ShamraniA. Artificial intelligence in paediatric emergencies: a narrative review. Am J Pediatr. 2022;8(2):51-55. doi:https://doi.org/10.11648/j.ajp.20220802.11
32.
Bartl-PokornyKDZittaCBeiritMVogrinecGSchullerBWPokornyFB. Focused review on artificial intelligence for disease detection in infants. Front Digit Health. 2024;6:1459640. doi:https://doi.org/10.3389/fdgth.2024.1459640
33.
Gómez-AndrésDOulhissaneAQuijano-RoyS. Two decades of advances in muscle imaging in children: from pattern recognition of muscle diseases to quantification and machine learning approaches. Neuromuscul Disord.2021;31(10):1038-1050. doi:https://doi.org/10.1016/j.nmd.2021.08.006
34.
AlqaralehMAl-BatahMEAMSA. Automated quantification of vesicoureteral reflux using machine learning with advancing diagnostic precision. Data Metadata. 2025;4:460. doi:https://doi.org/10.56294/dm2025756
35.
KhondkerAKwongJCRickardMSkretaMKeefeDTLorenzoAJErdmanL. A machine learning-based approach for quantitative grading of vesicoureteral reflux from voiding cystourethrograms: methods and proof of concept. J Pediatr Urol.2022;18(1):78.e1-78.e7. doi:https://doi.org/10.1016/j.jpurol.2021.10.009
36.
KhondkerAKwongJCYadavPChanJYHSinghASkretaMErdmanLKeefeDTFischerKTasianGHannickJHPapanikolaouFCooperBJCooperCSRickardMLorenzoAJ. Multi-institutional validation of improved vesicoureteral reflux assessment with simple and machine learning approaches. J Urol.2022;208(6):1314-1322. doi:https://doi.org/10.1097/JU.0000000000002987
37.
LiZTanZWangZTangWRenXFuJWangGChuHChenJDuanY, et al. Development and multi-institutional validation of a deep learning model for grading of vesicoureteral reflux on voiding cystourethrogram: a retrospective multicenter study. EClinicalMedicine. 2024;69:102466.
38.
ErgünOSerelTAÖztürkSASerelHBSoyupekSHoşcanB. Deep-learning-based diagnosis and grading of vesicoureteral reflux: a novel approach for improved clinical decision-making: vesicoureteral reflux and deep learning. J Surg Med. 2024;8(1):12-16.
39.
CampbellJPKimSJBrownJMOstmoSChanRVPKalpathy-CramerJChiangMFChiangMFOstmoSJin KimS, et al. Evaluation of a deep learning–derived quantitative retinopathy of prematurity severity scale. Ophthalmology. 2021;128(7):1070-1076. doi:https://doi.org/10.1016/j.ophtha.2020.10.025
40.
GuptaKCampbellJPTaylorSBrownJMOstmoSChanRVPDyJErdogmusDIoannidisSKalpathy-CramerJ, et al. A quantitative severity scale for retinopathy of prematurity using deep learning to monitor disease regression after treatment. JAMA Ophthalmol.2019;137(9):1029-1036. doi:https://doi.org/10.1001/jamaophthalmol.2019.2442
41.
TaylorSBrownJMGuptaKCampbellJPOstmoSChanRVPDyJErdogmusDIoannidisSKimSJet al. Monitoring disease progression with a quantitative severity scale for retinopathy of prematurity using deep learning. JAMA Ophthalmol.2019;137(9):1022-1028. doi:https://doi.org/10.1001/jamaophthalmol.2019.2433
42.
WuQHuYMoZZhangXYangYLiuBXiaoYZengXLinZFangY, et al. Development and validation of a deep learning model to predict the occurrence and severity of retinopathy of prematurity. JAMA Netw Open. 2022;5(6):e2217447-e2217447. doi:https://doi.org/10.1001/jamanetworkopen.2022.17447
43.
IqbalFSattiMIIrshadAShahMA. Predictive analytics in smart healthcare for child mortality prediction using a machine learning approach. Open Life Sci. 2023;18(1):20220609. doi:https://doi.org/10.1515/biol-2022-0609
44.
LépineCKleinPVoronTMandavitMBerrebiDOuth-GauerSPéréHTournierLPagèsFTartourE, et al. Histological severity risk factors identification in juvenile-onset recurrent respiratory papillomatosis: how immunohistochemistry and AI algorithms can help?Front Oncol.2021;11:596499.
45.
ForlanoRMullishBHGiannakeasNMauriceJBAngkathunyakulNLloydJTzallasATTsipourasMYeeMThurszMRet al. High-throughput, machine learning–based quantification of steatosis, inflammation, ballooning, and fibrosis in biopsies from patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol.2020;18(9):2081-2090.e9. doi:https://doi.org/10.1016/j.cgh.2019.12.025
46.
MizutaniKMiwaTSakamotoYTodaM. Application of deep learning techniques for automated diagnosis of non–syndromic craniosynostosis using skull. J Craniofac Surg. 2022;33(6):1843-1846. doi:https://doi.org/10.1097/SCS.0000000000008620
47.
GeislerELAgarwalSHallacRRDaescuOKaneAA. A role for artificial intelligence in the classification of craniofacial anomalies. J Craniofac Surg. 2021;32(3):967-969. doi:https://doi.org/10.1097/SCS.0000000000007369
48.
SabetiMBoostaniRMoradiEShakoorMH. Machine learning-based identification of craniosynostosis in newborns. Mach Learn Appl. 2022;8:100292. doi:ARTN10029210.1016/j.mlwa.2022.100292
49.
de JongGBijlsmaEMeulsteeJWennenMvan LindertEMaalTAquariusRDelyeH. Combining deep learning with 3D stereophotogrammetry for craniosynostosis diagnosis. Sci Rep.2020;10(1):15346. doi:https://doi.org/10.1038/s41598-020-72143-y
BlumJDBeirigerJVillavisanisDFMoralesCChoDYTaoWWhitakerRBartlettSPTaylorJAGoldsteinJASwansonJW. Machine learning in metopic craniosynostosis: does phenotypic severity predict long-term esthetic outcome?J Craniofac Surg. 2023;34(1):58-64. doi:https://doi.org/10.1097/SCS.0000000000008868
52.
BlumJDBeirigerJKalmarCAveryRALangS-SVillavisanisDFCheungLChoDYTaoWWhitakerR, et al.Relating metopic craniosynostosis severity to intracranial pressure. J Craniofac Surg. 2022;33(8):2372-2378. doi:https://doi.org/10.1097/SCS.0000000000008748
53.
BeirigerJWTaoWIrgebayZSmetonaJDvoracekLKassNMDixonAZhangCMehtaMWhitakerR, et al.A longitudinal analysis of pre-and post-operative dysmorphology in metopic craniosynostosis. Cleft Palate Craniofac J. 2024. doi:https://doi.org/10.1177/10556656241237605
54.
RomeoDJBecerraSLBarreroCEMassenburgBBDuSWuMSwansonJWTaylorJABartlettSP. Validating the supraorbital notch-nasion-supraorbital notch angle as a marker for metopic craniosynostosis severity. J Craniofac Surg. 2024;35(4):1035-1039.
55.
SchaufelbergerMKühleRWachterAWeichelFHagenNRingwaldFEisenmannUHoffmannJEngelMFreudlspergerCNahmW. A radiation-free classification pipeline for craniosynostosis using statistical shape modeling. Diagnostics. 2022;12(7):1516. doi:https://doi.org/10.3390/diagnostics12071516
56.
PorrasARTuLTseringDMantillaEOhAEnquobahrieAKeatingRRogersGFLinguraruMG. Quantification of head shape from three-dimensional photography for presurgical and postsurgical evaluation of craniosynostosis. Plast Reconstr Surg.2019;144(6):1051e-1060e. doi:https://doi.org/10.1097/PRS.0000000000006260
57.
SaarikkoAMellanenEKuuselaLLeikolaJKarppinenAAuttiTVirtanenPBrandstackN. Comparison of black bone MRI and 3D-CT in the preoperative evaluation of patients with craniosynostosis. J Plast Reconstr Aesthet Surg.2020;73(4):723-731.
