Neural network classification of pediatric posterior fossa tumors using clinical and imaging data

Pediatr Neurosurg. Jan-Feb 2004;40(1):8-15. doi: 10.1159/000076571.

Abstract

A neural network was developed that utilizes both clinical and imaging (CT and MRI) data to predict posterior fossa tumor (PFT) type. Data from 33 children with PFTs were used to develop and test the system. When all desired information was available, the network was able to correctly classify 85.7% of the tumors. In cases with incomplete data, it was able to correctly classify 72.7% of the tumors. In both instances, the diagnoses made by the network were more likely to be correct than those made by the neuroradiologists.

MeSH terms

  • Astrocytoma / diagnostic imaging*
  • Astrocytoma / pathology*
  • Cerebellar Neoplasms / diagnostic imaging*
  • Cerebellar Neoplasms / pathology*
  • Child
  • Cranial Fossa, Posterior / diagnostic imaging*
  • Cranial Fossa, Posterior / pathology*
  • Ependymoma / diagnostic imaging*
  • Ependymoma / pathology*
  • Female
  • Humans
  • Magnetic Resonance Imaging
  • Male
  • Medulloblastoma / pathology*
  • Models, Neurological
  • Neural Networks, Computer*
  • Skull Neoplasms / diagnostic imaging*
  • Skull Neoplasms / pathology*
  • Tomography, X-Ray Computed