A Region-Based Ontology of the Brain Ventricular System and Its Relation to Schizophrenia

A Region-Based Ontology of the Brain Ventricular System and Its Relation to Schizophrenia

Paulo Santos (Centro Universitário da FEI, Brazil), Rodolpho Freire (Centro Universitário da FEI, Brazil), Danilo N. dos Santos (Centro Universitário da FEI, Brazil), Carlos Thomaz (Centro Universitário da FEI, Brazil), Paulo Sallet (Universidade de Sao Paulo, Brazil), Mario Louzã (Universidade de Sao Paulo, Brazil) and Anthony G. Cohn (University of Leeds, UK)
DOI: 10.4018/978-1-61692-868-1.ch007


This chapter describes an initial region-based formalisation of some concepts about neuroanatomy into ontological and epistemic terms, as part of a major effort into the formalisation of the knowledge contained in neuroimages of patients with schizophrenia. The long-term goal is to build an ontology that is a formal basis for the expectations generated from statistical data analysis. To this end, the chapter presents an example of applying this ontology to interpret the results of image-based analysis of neuroimages from schizophrenic patients.
Chapter Preview

1. Introduction

Schizophrenia is a mental disorder that affects approximately 1% of the population worldwide and whose causes are still unknown1. Both genetic (Harrison & Owen, 2003) and environmental factors (including biological—e.g., prenatal infection and obstetric complications—and psychosocial factors) appear to play a role in its etiology. These factors, however, are not sufficient for the emergence of schizophrenia (Nuechterlein & Dawson, 1984). Previous studies (Gattaz, et al., 1988; Thomaz, et al., 2007b) investigated the possibility that the neuroanatomical changes of this disease are diffuse, therefore suggesting that multidimensional techniques may be the right tools for providing information of the possible influence of structural brain anomalies related to the disease. In another work (Santos, et al., 2009), we have proposed an integrated framework for extracting and describing patterns of changes from neuroimages of schizophrenic patients using a combination of linear discriminant analysis and active contour models. As a result, we obtained clusters of the most statistically significant differences of neuroimages between healthy controls and patients (an example of such clusters is shown in Figure 1).

Figure 1.

Results of the active contour model segmentation from statistical

The voxels within these clusters were mapped using the Talairach atlas (Lancaster, et al., 2000) to the corresponding labels of neuroanatomic structures. However, that work falls short on placing the obtained descriptions into the context of neuroanatomy, as the relationships between the labels and the structures they describe were not represented explicitly. Besides, the knowledge about the known facts of how anatomical changes are related to schizophrenia was also absent.

Complete Chapter List

Search this Book: