Machine learning applied to neuroimaging for diagnosis of adult classic Chiari malformation: role of the basion as a key morphometric indicator

Restricted access


The current diagnostic criterion for Chiari malformation Type I (CM-I), based on tonsillar herniation (TH), includes a diversity of patients with amygdalar descent that may be caused by a variety of factors. In contrast, patients presenting with an overcrowded posterior cranial fossa, a key characteristic of the disease, may remain misdiagnosed if they have little or no TH. The objective of the present study was to use machine-learning classification methods to identify morphometric measures that help discern patients with classic CM-I to improve diagnosis and treatment and provide insight into the etiology of the disease.


Fifteen morphometric measurements of the posterior cranial fossa were performed on midsagittal T1-weighted MR images obtained in 195 adult patients diagnosed with CM. Seven different machine-learning classification methods were applied to images from 117 patients with classic CM-I and 50 controls matched by age and sex to identify the best classifiers discriminating the 2 cohorts with the minimum number of parameters. These classifiers were then tested using independent CM cohorts representing different entities of the disease.


Machine learning identified combinations of 2 and 3 morphometric measurements that were able to discern not only classic CM-I (with more than 5 mm TH) but also other entities such as classic CM-I with moderate TH and CM Type 1.5 (CM-1.5), with high accuracy (> 87%) and independent of the TH criterion. In contrast, lower accuracy was obtained in patients with CM Type 0. The distances from the lower aspect of the corpus callosum, pons, and fastigium to the foramen magnum and the basal and Wackenheim angles were identified as the most relevant morphometric traits to differentiate these patients. The stronger significance (p < 0.01) of the correlations with the clivus length, compared with the supraoccipital length, suggests that these 5 relevant traits would be affected more by the relative position of the basion than the opisthion.


Tonsillar herniation as a unique criterion is insufficient for radiographic diagnosis of CM-I, which can be improved by considering the basion position. The position of the basion was altered in different entities of CM, including classic CM-I, classic CM-I with moderate TH, and CM-1.5. The authors propose a predictive model based on 3 parameters, all related to the basion location, to discern classic CM-I with 90% accuracy and suggest considering the anterior alterations in the evaluation of surgical procedures and outcomes.

ABBREVIATIONS CM = Chiari malformation; DT = decision tree; FM = foramen magnum; k-NN = k-nearest neighbors; LR = logistic regression; NB = naïve Bayes; PCF = posterior cranial fossa; SVM = support vector machine; TH = tonsillar herniation.

Article Information

Correspondence Aintzane Urbizu, Duke Molecular Physiology Institute, Duke University School of Medicine, 300 N Duke St., Durham, NC 27701. email:

INCLUDE WHEN CITING Published online October 20, 2017; DOI: 10.3171/2017.3.JNS162479.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

© AANS, except where prohibited by US copyright law.



  • View in gallery

    Representative sagittal T1-weighted MR images corresponding to each of the analyzed cohorts: controls (A), CM-I (B and C), CM-0 (D), CM-I with moderate TH (E), and CM-1.5 (F). Numbers indicate the measurements performed in all individuals: 1 indicates tentorium length; 2, supraoccipital length; 3, anteroposterior diameter of FM; 4, clivus length; 5, PCF area; 6, osseous PCF area; 7, tentorium angle; 8, height PCF; 9, width of PCF; 10, basilar impression; 11, distance from pons to FM; 12, distance from corpus callosum to FM; 13, distance from fastigium to FM; 14, Wackenheim angle; and 15, basal angle. Panels B and C are modified from Urbizu A, Poca MA, Vidal X, Rovira A, Sahuquillo J, Macaya A: MRI-based morphometric analysis of posterior cranial fossa in the diagnosis of Chiari malformation type I. J Neuroimaging 24:250–256, 2014. Used with permission. Figure is available in color online only.

  • View in gallery

    Plots showing analyses of correlation between the clivus length and the 5 measurements most relevant according to the 7 supervised classification methods—the distance from the pons, the corpus callosum (CC), and the fastigium to the FM; the basal angle; and the Wackenheim angle (A–E, respectively)—and between the distance from the pons to the FM and the basal angle (F). Blue dots indicate measurements obtained in controls (n = 50), and green dots those obtained in patients with CM-I (n = 117). The respective correlation coefficients and p values are shown for each analysis. Figure is available in color online only.

  • View in gallery

    Ordination plots depicting the distribution of data points used for training (left) and validation (right). The classification line generated by the support vector machine using only 2 features is also shown. Figure is available in color online only.



Aboulezz AOSartor KGeyer CAGado MH: Position of cerebellar tonsils in the normal population and in patients with Chiari malformation: a quantitative approach with MR imaging. J Comput Assist Tomogr 9:103310361985


Aitken LALindan CESidney SGupta NBarkovich AJSorel M: Chiari type I malformation in a pediatric population. Pediatr Neurol 40:4494542009


Alperin NLoftus JROliu CJBagci ALee SHErtl-Wagner B: Magnetic resonance imaging measures of posterior cranial fossa morphology and cerebrospinal fluid physiology in Chiari malformation type I. Neurosurgery 75:5155222014


Aydin SHanimoglu HTanriverdi TYentur EKaynar MY: Chiari type I malformations in adults: a morphometric analysis of the posterior cranial fossa. Surg Neurol 64:2372412005


Barkovich AJWippold FJSherman JLCitrin CM: Significance of cerebellar tonsillar position on MR. AJNR Am J Neuroradiol 7:7957991986


Bizios DHeijl AHougaard JLBengtsson B: Machine learning classifiers for glaucoma diagnosis based on classification of retinal nerve fibre layer thickness parameters measured by Stratus OCT. Acta Ophthalmol 88:44522010


Breiman LFriedman JHOlshen RAStone CJ: Classification and Regression Trees. Boca Raton, FL: Chapman & Hall1984


Cesmebasi ALoukas MHogan EKralovic STubbs RSCohen-Gadol AA: The Chiari malformations: a review with emphasis on anatomical traits. Clin Anat 28:1841942015


Chiari H: Concerning alterations in the cerebellum resulting from cerebral hydrocephalus. 1891. Pediatr Neurosci 13:381987


Chiari H: Über Veränderungen des Kleinhirns, des Pons und der Medulla Oblongata infolge von congenitaler Hydrocephalie des Grosshirns. Denkschr der Kais Akad Wiss Wien math Naturw 63:711161895


Chotai SMedhkour A: Surgical outcomes after posterior fossa decompression with and without duraplasty in Chiari malformation-I. Clin Neurol Neurosurg 125:1821882014


Dagtekin AAvci EKara EUzmansel DDagtekin OKoseoglu A: Posterior cranial fossa morphometry in symptomatic adult Chiari I malformation patients: comparative clinical and anatomical study. Clin Neurol Neurosurg 113:3994032011


Deng XYang CGan JYang BWu LYang T: Chiari malformation type 1.5 in male monozygotic twins: case report and literature review. Clin Neurol Neurosurg 130:1551582015


Dobson AJBarnett A: An Introduction to Generalized Linear Modelsed 3. Boca Raton, FL: CRC Press2008


Fix EHodges JL Jr: Discriminatory Analysis. Nonparametric Discrimination: Consistency Properties. Randolph Field, TX: USAF School of Aviation Medicine1951


Förander PSjåvik KSolheim ORiphagen IGulati SSalvesen Ø: The case for duraplasty in adults undergoing posterior fossa decompression for Chiari I malformation: a systematic review and meta-analysis of observational studies. Clin Neurol Neurosurg 125:58642014


Garrard PRentoumi VGesierich BMiller BGorno-Tempini ML: Machine learning approaches to diagnosis and laterality effects in semantic dementia discourse. Cortex 55:1221292014


Greenlee JDDonovan KAHasan DMMenezes AH: Chiari I malformation in the very young child: the spectrum of presentations and experience in 31 children under age 6 years. Pediatrics 110:121212192002


Guerreiro RBBittencourt LReis RCRotta JMTufik SBotelho RV: Upper airway dimensions in patients with craniocervical junction malformations with and without sleep apnea. A pilot case-control study. Arq Neuropsiquiatr 73:3363412015


Halvorson KGKellogg RTKeachie KNGrant GAMuh CRWaldau B: Morphometric Analysis of Predictors of Cervical Syrinx Formation in the Setting of Chiari I Malformation. Pediatr Neurosurg 51:1371412016


Isik NElmaci IKaksi MGokben BIsik NCelik M: A new entity: Chiari Zero malformation and its surgical method. Turk Neurosurg 21:2642682011


Iskandar BJHedlund GLGrabb PAOakes WJ: The resolution of syringohydromyelia without hindbrain herniation after posterior fossa decompression. J Neurosurg 89:2122161998


Karagöz FIzgi NKapíjcíjoğlu Sencer S: Morphometric measurements of the cranium in patients with Chiari type I malformation and comparison with the normal population. Acta Neurochir (Wien) 144:1651712002


Khan AABhatti SNKhan GAhmed EAurangzeb AAli A: Clinical and radiological findings in Arnold Chiari malformation. J Ayub Med Coll Abbottabad 22:75782010


Khazaee AEbrahimzadeh ABabajani-Feremi A: Application of advanced machine learning methods on resting-state fMRI network for identification of mild cognitive impairment and Alzheimer’s disease. Brain Imaging Behav 10:7998172016


Kim IKWang KCKim IOCho BK: Chiari 1.5 malformation: an advanced form of Chiari I malformation. J Korean Neurosurg Soc 48:3753792010


Krishan KKanchan T: Evaluation of spheno-occipital synchondrosis: A review of literature and considerations from forensic anthropologic point of view. J Forensic Dent Sci 5:72762013


Manning CDRaghavan PSchutze H: Introduction to Information Retrieval. Cambridge: Cambridge University Press2008Vol 1


Marin-Padilla MMarin-Padilla TM: Morphogenesis of experimentally induced Arnold–Chiari malformation. J Neurol Sci 50:29551981


Markunas CATubbs RSMoftakhar RAshley-Koch AEGregory SGOakes WJ: Clinical, radiological, and genetic similarities between patients with Chiari Type I and Type 0 malformations. J Neurosurg Pediatr 9:3723782012


Meadows JKraut MGuarnieri MHaroun RICarson BS: Asymptomatic Chiari Type I malformations identified on magnetic resonance imaging. J Neurosurg 92:9209262000


Milhorat THChou MWTrinidad EMKula RWMandell MWolpert C: Chiari I malformation redefined: clinical and radiographic findings for 364 symptomatic patients. Neurosurgery 44:100510171999


Milhorat THNishikawa MKula RWDlugacz YD: Mechanisms of cerebellar tonsil herniation in patients with Chiari malformations as guide to clinical management. Acta Neurochir (Wien) 152:111711272010


Moncho DPoca MAMinoves TFerré ACañas VSahuquillo J: Are evoked potentials clinically useful in the study of patients with Chiari malformation Type 1? J Neurosurg 126:6066192017


Mottolese CSzathmari ASimon ERousselle CRicci-Franchi ACHermier M: Treatment of Chiari type I malformation in children: the experience of Lyon. Neurol Sci 32 (Suppl 3):S325S3302011


Mueller DMOro’ JJ: Prospective analysis of presenting symptoms among 265 patients with radiographic evidence of Chiari malformation type I with or without syringomyelia. J Am Acad Nurse Pract 16:1341382004


Nishikawa MSakamoto HHakuba ANakanishi NInoue Y: Pathogenesis of Chiari malformation: a morphometric study of the posterior cranial fossa. J Neurosurg 86:40471997


Noudel RGomis PSotoares GBazin APierot LPruvo JP: Posterior fossa volume increase after surgery for Chiari malformation Type I: a quantitative assessment using magnetic resonance imaging and correlations with the treatment response. J Neurosurg 115:6476582011


Noudel RJovenin NEap CScherpereel BPierot LRousseaux P: Incidence of basioccipital hypoplasia in Chiari malformation type I: comparative morphometric study of the posterior cranial fossa. Clinical article. J Neurosurg 111:104610522009


Pang DThompson DN: Embryology and bony malformations of the craniovertebral junction. Childs Nerv Syst 27:5235642011


Sekula RF JrJannetta PJCasey KFMarchan EMSekula LKMcCrady CS: Dimensions of the posterior fossa in patients symptomatic for Chiari I malformation but without cerebellar tonsillar descent. Cerebrospinal Fluid Res 2:112005


Sgouros SKountouri MNatarajan K: Posterior fossa volume in children with Chiari malformation Type I. J Neurosurg 105 (2 Suppl):1011062006


Shehu BBIsmail NJMahmud MRHassan I: Chiari I malformation: a missed diagnosis. Ann Afr Med 5:2062082006


Silva FRVidotti VGCremasco FDias MGomi ESCosta VP: Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using spectral domain OCT and standard automated perimetry. Arq Bras Oftalmol 76:1701742013


Stone M: Cross-validatory choice and assessment of statistical predictions. J R Stat Soc B 36:1111471974


Strahle JMuraszko KMKapurch JBapuraj JRGarton HJMaher CO: Chiari malformation Type I and syrinx in children undergoing magnetic resonance imaging. J Neurosurg Pediatr 8:2052132011


Tubbs RSElton SGrabb PDockery SEBartolucci AAOakes WJ: Analysis of the posterior fossa in children with the Chiari 0 malformation. Neurosurgery 48:105010552001


Tubbs RSIskandar BJBartolucci AAOakes WJ: A critical analysis of the Chiari 1.5 malformation. J Neurosurg 101 (2 Suppl):1791832004


Tubbs RSOakes WJ (eds): The Chiari Malformations. New York: Springer2013


Urbizu AFerre APoca MARovira ASahuquillo JMartin BA: Cephalometric oropharynx and oral cavity analysis in Chiari malformation Type I: a retrospective case-control study. J Neurosurg 126:6266332016


Urbizu APoca MAVidal XRovira ASahuquillo JMacaya A: MRI-based morphometric analysis of posterior cranial fossa in the diagnosis of Chiari malformation type I. J Neuroimaging 24:2502562014


Urbizu AToma CPoca MASahuquillo JCuenca-León ECormand B: Chiari malformation type I: a case-control association study of 58 developmental genes. PLoS One 8:e572412013


Vapnik VN: The Nature of Statistical Learning Theory. New York: Springer2013


Witten IHFrank EHall M: Data Mining: Practical Machine Learning Tools and Techniquesed 3. Burlington, MA: Morgan Kaufmann Publishers2011


Yan HHan XJin MLiu ZXie DSha S: Morphometric features of posterior cranial fossa are different between Chiari I malformation with and without syringomyelia. Eur Spine J 25:220222092016


Zhao JLLi MHWang CLMeng W: A systematic review of Chiari I malformation: techniques and outcomes. World Neurosurg 88:7142016




All Time Past Year Past 30 Days
Abstract Views 102 102 44
Full Text Views 230 230 36
PDF Downloads 177 177 12
EPUB Downloads 0 0 0


Google Scholar