First Positron Emission Tomography at Brookhaven

01/01/1961View on timeline

Positron-emission tomography (PET) is a nuclear medicine functional imaging technique that is used to observe metabolic processes in the body as an aid to the diagnosis of disease. The system detects pairs of gamma rays emitted indirectly by a positron-emitting radionuclide, most commonly fluorine-18. which is introduced into the body on a biologically active molecule called a radioactive tracer. Three-dimensional images of tracer concentration within the body are then constructed by computer analysis. In modern PET-CT scanners, three-dimensional imaging is often accomplished with the aid of a CT X-ray scan performed on the patient during the same session, in the same machine.

If the biologically active tracer molecule chosen for PET is fludeoxyglucose (FDG), an analogue of glucose, the concentrations of tracer imaged will indicate tissue metabolic activity as it corresponds to the regional glucose uptake. Use of this tracer to explore the possibility of cancer metastasis (i.e., spreading to other sites) is the most common type of PET scan in standard medical care (representing 90% of current scans). Metabolic trapping of the radioactive glucose molecule allows the PET scan to be utilized. The same tracer may also be used for PET investigation and diagnosis of types of dementia. Less often, other radioactive tracers, usually but not always labeled with fluorine-18, are used to image the tissue concentration of other types of molecules of interest.

In 1961, James Robertson et al. at the famous Brookhaven National Laboratory built the first single-plane PET device and gave it a funny nickname of the “head-shrinker.”. It was then realized that there was no reason to confine the scanning detectors to a plane and an arrangement whereby 32 detectors were arranged on the surface of a sphere was tried. There was then no need for movement in order to generate a three-dimensional distribution. The device was built and christened the ‘non-inertial positron scanner’ although it was known locally by some as ‘hair drier’.

An approach was abandoned when it became clear that 32 crystals in this arrangement could not give the required resolution and that data from a 32-crystal system, let alone one with more detectors, was too complicated to analyze. It was therefore decided to work with the 32 crystals arranged in a single plane. The system was designed for brain investigations.

The Brookhaven prototype 32-crystal ring system positron camera - the 'head shrinker'.

The ring of 32 crystals was mounted on a circle 40 cm in diameter. The crystals were sodium iodide 3.2 cm in diameter. Each crystal detector was electronically coupled for potential coincidence with each of its 31 neighbors. For example with reference to figure below, if a point source of emission was located at the darkened spot, the arrowed lines indicate the possible back-to-back photon paths which would lead to coincidence detection. The patient was seated on a dental chair for positioning. This machine became affectionately known as the ‘head shrinker’. It was constructed in 1961 and very quickly replaced the role of the ‘hair drier’.

The principle of operation of a ring-system positron camera.

From the coincidence data recorded, a planar section through the head can in principle be reconstructed. Essentially the coincidence counts on some particular channel (that is a pair of detectors) must relate to the sum of activity along the line joining the detectors with appropriate weighting for the geometric efficiency.

The equipment development began early in the 1960s but there were problems with the reconstruction. The first methods considered were matrix inversion techniques but these were not found to be very satisfactory. Certain simplifications were then adopted. It is inherently obvious that, if coincidence channels between detectors with fixed separations around the circumference of the circle were considered, then these channels are insensitive to activity within a circle defined by the radius of a chord connecting any particular pair. For detectors with small circumferential separation is increased, the circle of exclusion decreases in radius until it is just the central point for detectors at opposite ends of a diameter. This led to the development of an ‘onion-peeling’ algorithm for reconstruction.