Abstract
An important goal of software development in the medical field is the design of methods which are able to integrate information obtained from various imaging and nonimaging modalities into a cohesive framework in order to understand the results of qualitatively different measurements in a larger context. Moreover, it is essential to assess the various features of the data quantitatively so that relationships in anatomical and functional domains between complementing modalities can be expressed mathematically. This paper presents a clinically feasible software environment for the quantitative assessment of the relationship among biochemical functions as assessed by PET imaging and electrophysiological parameters derived from intracranial EEG. Based on the developed software tools, quantitative results obtained from individual modalities can be merged into a data structure allowing a consistent framework for advanced data mining techniques and 3D visualization. Moreover, an effort was made to derive quantitative variables (such as the spatial proximity index, SPI) characterizing the relationship between complementing modalities on a more generic level as a prerequisite for efficient data mining strategies. We describe the implementation of this software environment in twelve children (mean age 5.2 ± 4.3 years) with medically intractable partial epilepsy who underwent both high-resolution structural MR and functional PET imaging. Our experiments demonstrate that our approach will lead to a better understanding of the mechanisms of epileptogenesis and might ultimately have an impact on treatment. Moreover, our software environment holds promise to be useful in many other neurological disorders, where integration of multimodality data is crucial for a better understanding of the underlying disease mechanisms.
Highlights
With ever-improving imaging technologies and high-performance computational power, the complexity and scale of brain imaging data have continued to grow at an explosive pace
Recent advances in imaging technologies, such as molecular imaging using positron emission tomography (PET), structural imaging using high-resolution magnetic resonance (MR), as well as quantitative electrophysiological cortical mapping using electroencephalography (EEG), have allowed an increased understanding of normal and abnormal brain structures and functions [1,2,3,4] It is well understood that normal brain function is dependent on the interactions between specialized regions of the brain which process information within local and global networks
We separated the computational components from other functionality and created a standalone programming library consisting of the following modules: (1) conformal mapping; (2) objective definition of PET abnormalities; (3) EEG grid overlay; (4) calculation of spatial proximity index (SPI) values; (5) extendable database component; and (6) graphical output module
Summary
With ever-improving imaging technologies and high-performance computational power, the complexity and scale of brain imaging data have continued to grow at an explosive pace. In addition to integration, advanced computational tools need to be developed which allow quantitative analysis of a variety of functional patterns as well as the design of a software environment that allows quantitative assessment of relationships between diverse functional and anatomical features. This is an area of active research, current state-of-the-art technologies are still suboptimal with respect to multimodality integration and quantitative assessment of qualitatively distinct neuroimaging datasets in patients with structurally abnormal brains. This paper presents a multimodality database for neuroimaging, electrophysiological, as well as clinical data designed for the assessment of epileptic foci.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
