E-Health Decision Support Systems for the Diagnosis of Dementia Diseases

E-Health Decision Support Systems for the Diagnosis of Dementia Diseases

Isabella Castiglioni (Institute of Molecular Bioimaging and Physiology, National Research Council, Italy), Maria Carla Gilardi (Institute of Molecular Bioimaging and Physiology, National Research Council, Italy) and Francesca Gallivanone (Institute of Molecular Bioimaging and Physiology, National Research Council, Italy)
DOI: 10.4018/978-1-4666-2657-7.ch006
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The increase of incidence and prevalence of dementia diseases makes urgent the clinical community to be supported in the difficult diagnostic process of dementia patients. E-health decision support systems, based on innovative algorithms able to extract information from in vivo neuroimaging studies, can make a quite different way to perform neurological diagnosis and enlarge domains and actors involved in the diagnostic process. A number of image-processing methods that extract potential biomarkers from the in vivo neuroimaging studies have been proposed (e.g. volume segmentation, voxel-based statistical mapping). A number of new shape descriptors have also been developed (e.g. texture-based). Other approaches (e.g. machine learning, pattern recognition) have been proven effective, for both structural and functional data, in making automatic diagnoses. The integration of these sophisticated diagnostic tools into secure, efficient, and wide e-infrastructures is the prerequisite for the real implementation of e-health support services to the clinical and industrial communities managing dementia patients.
Chapter Preview
Top

Background

The first clinical criteria for the Alzheimer disease (AD) (1984) defined it as having a single stage, with diagnosis based on clinical symptoms. According to these criteria, disease-free subjects were considered dementia-free subjects and the only proof of AD was represented by autopsy, through the presence of deposited amyloid plaques and τ protein tangles.

Scientific evidences coming from in neuroimaging studies of patients with suspected dementia diseases showed the potential of in vivo non invasive imaging technologies to support medical decisions in the diagnosis of dementia diseases.

In 1994, The Quality Standards Subcommittee of the American Academy of Neurology published “practice parameters” for the diagnostic evaluation of persons suspected of having dementia. These guidelines established neurologic examination as the standard, and certain laboratory tests as essential components of the diagnostic procedure. However, neuroimaging studies were suggested only as “options,” i.e. “management strategies for which there is unclear clinical certainty.”

A few years later, in 1997, the usefulness of these practice parameters for the assessment of dementia was examined by reviewing a large number of consecutive cases. The main finding of these evaluations was that diagnostic accuracy was improved by neuroimaging studies, although with a predictable increase in costs. Findings proved that neuroimaging studies frequently changed clinical diagnosis (19%-28%) and patient management (15%).

Structural neuroimaging techniques, such as Computerized Tomography (CT) and Magnetic Resonance Imaging (MRI), provided useful information in the diagnosis of AD and in predicting the development of dementia, e.g. by measurements of hippocampus atrophy, where the most prominent structural damage occurs at a macroscopic level.

Complete Chapter List

Search this Book:
Reset