Volume Assessment of the Cerebrospinal Fluid Spaces for Computer Aided Diagnosis

Alain Lebret (1) , Eric Petit (1) , Bruno Durning (1) , Jérôme Hodel (2) , Philippe Decq (3) , Alain Rahmouni (2)

1 LISSI - Laboratoire Images, Signaux et Systèmes Intelligents
2 Service d'imagerie médicale
3 Service de neurochirurgie [CHU Créteil]

The present work aims to provide support in the diagnosis of hydrocephalus, which requires an assessment to the volumes of the cerebrospinal fluid (CSF) within its total, ventricular and subarachnoid spaces. In this paper we describe a fully automatic method to estimate the CSF volumes from a new 3D whole body MR imaging sequence. The method was developed using image properties as well as anatomical and geometrical features, completed with a topological assumption on the CSF shape. Experiments on phantoms and clinical data were performed and evaluated by comparing our assessments of volumes with those derived from a segmentation controlled by expert physicians. Then we show that a robust distinction between pathological cases and healthy adult people can be achieved by a linear discriminant analysis on volumes of the ventricular and intracranial subarachnoid spaces. We also find that healthy adults maintain a proportional relationship between these volumes.

Mots clés: cerebrospinal fluid; hydrocephalus; 3D segmentation; volume; CSF; computer-aided diagnosis; liquide cérébrospinal; LCR; LCS

Domaines: Imagerie médicale; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Alain Lebret, Eric Petit, Bruno Durning, Jérôme Hodel, Philippe Decq, et al.. Volume Assessment of the Cerebrospinal Fluid Spaces for Computer Aided Diagnosis. 2011. ⟨hal-00638369⟩