Peter J. Ell

Models of Brain Processing with Respect to Functional Neuroimaging

The cerebral cortex contains approximately 1014 synapses (von der Malsburg and Singer 1988). Current neuroimaging techniques can only resolve brain regions containing millions of cells. Hence it is always the coordinated functioning of millions of synapses and neurons that generate measurable activity. A single neuron cannot alone accomplish the processing from which complex behaviours emerge. Thus, in attempting to understand the neuronal basis of complex behaviour it is essential to have a theoretical model of the interactions of large groups of functional units. This not only makes data management convenient; it also in some way represents biological reality.

Clinical Use of SPET Imaging in Psychiatric and Neurological Disease

Recent advances in SPET technology have produced increased sensitivity and resolution of SPET systems (see Chaps. 2 and 3), allowing the technique to be applied in new and exciting ways to the understanding of clinical disease, particularly in the expanding fields of neurology and psychiatry. Functional neuroimaging has allowed clinicians to ‘bypass the skull’ and examine the active human brain in health and disease. SPET can aid doctors as never before as they make complex clinical decisions about patients.

A Historical and Philosophical Review of Localizing Brain Function

Functional neuroimaging with SPET and PET can be viewed as the latest chapter in a long historical debate about how the mind interacts with the brain. This debate is a fundamental one for mankind, as we try to understand how we are able to perform those actions that most make us human — thinking, planning, dreaming, grieving. It is both interesting and entertaining to look back over this debate about how the brain and mind work. However, reviewing this history is more than simply an amusing historical game. For only by looking back and realizing how we arrived at our present theories of how the brain functions can we then proceed to ask the important questions that functional neuroimaging allows us to investigate.

SPET Instrumentation: Characteristics and New Developments

According to the principles of emission imaging outlined in Chap. 1, a single-photon emission tomography (SPET) system must be able to detect gamma rays along a multitude of directions. How the transition is made from infinite sampling, as required by the mathematical theory (Radon 1917; Herman 1980), to finite sampling in practice is critical (Budinger 1980; Heller and Goodwin 1987). Therefore, the type and performance of different SPET systems are dependent on how the various directions are physically realized and how well they sample the region to be reconstructed.

Evaluation of Ventricular Function

Modern views of circulatory physiology see the heart and peripheral vasculature as a closely integrated system in which the heart acts as a pump, supplying the energy, with small contributions from peripheral and respiratory musculature, for the circulation of blood (1, 2). The cardiac output is influenced by a number of variables which can be thought of as cardiac or vascular, i.e., blood volume and viscosity, peripheral resistance, venomotor tone and vascular compliance. In turn, the cardiac factors can be thought of as those affecting ‘pump performance’ and others, such as the pattern of electrical activation, valvar stenosis or regurgitation, the presence of regional abnormalities of ventricular contraction or intracardiac shunts and ventricular compliance. By analogy with the contraction of isolated muscle (3), the variables affecting ‘pump performance’ are widely held to be heart rate, preload (tension before contraction begins), afterload (tension during contraction) and a variable thought to be independent of load and referred to as contractility.

Description and Principles of Neuroactivation

Recent advances in our knowledge of the local patterns of brain functional activity began with the studies of Lassen (Lassen et al. 1978) who developed techniques to measure blood flow in circumscribed regions of the intact human brain with the aid of radioactive isotopes. Even in the early studies, which required carotid artery cannulation and which had poor resolution by modern standards, Lassen and his colleagues demonstrated specific repeatable patterns of cortical blood flow associated with particular states and tasks. More recently the development of positron emission tomography (PET) and single-photon emission tomography (SPET) has further revolutionized our ability to view the living, working human brain.

Principles of Emission Tomography

Medical radionuclide imaging involves the administration to a patient of an appropriate radionuclide linked to a specific pharmaceutical. Most radionuclides used for medical imaging purposes fall into two broad categories (Tables 1 and 2): (a) Single photon emitters: radionuclides that predominantly emit gamma or X-rays with energies usually varying between 80 and 360 keV; (b) positron emitters: radionuclides that emit positrons which essentially behave like positive electrons. After colliding with neighbouring electrons, thereby losing its kinetic energy, the positron (now at rest or almost at rest) unites with an electron in a process called positron-electron annihilation, which results in the emission of two gamma rays of 511 keV energy, travelling in almost opposite directions. Whether with single photon or with positron emitters, the imaging process relies on the detection of emitted gamma rays. A position- and energy-sensitive radiation detector is used, coupled to a computer system which controls the data acquisition and performs the data processing.

Activation Studies in Specific Disease States: Future Applications

In Chap. 9 we reviewed the exciting potential of SPET neuroactivation for understanding how the brain works in healthy controls. Despite this potential for studying normal brain function, SPET may possibly lag behind PET in this area over the next decade for several reasons. First, it is usually easier to label biologically active compounds with a radioactive carbon, nitrogen, fluoride or oxygen atom than it is to label them with single photon emitters such as 99mTc or 123I. Additionally, the shorter physical half-life of the tracers used in PET allow multiple images (6–10) to be taken of the same subject in a single activation session, thus increasing the statistical power to detect subtle functional changes. Of course, these are countered by the expense and relative non-availability of PET as we saw in Chaps. 1 and 2.

Review of Previous SPET Neuroactivation Studies in Normal Controls

There are as yet very few SPET activation studies in the published literature. In part, this is because in the first instance functional neuroimaging was used to image the human brain at rest in health and disease. It was only when the technique was familiar and when the appearance of the brain at rest was known that attempts were made to investigate how these appearances changed when the brain received some sort of additional stimulus. Additionally, activation studies make greater demands on the versatility of radioactive ligands and data processing, and until recently most research efforts in these fields have been applied to positron emission tomography (PET).

Radiopharmaceutical Ligands and Tracers

In recent years improvements in instrumentation and radiopharmaceutical developments have broadened the scope of single-photon emission tomography (SPET) of the brain. Patients with neurological and psychiatric diseases may be better assessed. More objective parameters (i.e. brain perfusion, neuroreceptor availability, neurotransmitter utilization) can now be used to study the natural progression of disease and to determine its response to either established or new therapeutic interventions.