A different kind of image
Most imaging methods show structure. PET, or positron emission tomography, shows where a chemical goes in the body. A patient receives a very small amount of a compound labeled with a radioactive atom, and the scanner maps where the signal comes from. The method owes as much to chemistry and nuclear physics as to engineering.
Tracers: the older idea
The tracer principle goes back to the Hungarian chemist George de Hevesy, who in the 1910s used radioactive atoms to follow substances through chemical and biological systems. He received the 1943 Nobel Prize in Chemistry. The 1930s brought the cyclotron, developed by Ernest Lawrence, and the discovery of artificial radioactivity by Irène and Frédéric Joliot-Curie, which allowed scientists to make new isotopes on demand. Doctors began using some of these in the late 1930s and 1940s, for example with the thyroid.
Positrons and coincidence
Carl Anderson identified the positron, the electron’s antimatter partner, in 1932. When a positron is emitted from an unstable atom, it travels only a short distance before meeting an electron. Both vanish and two gamma rays leave in nearly opposite directions. That detail is the basis for PET. If two detectors on opposite sides of the body register a gamma ray at almost the same moment, the decay happened somewhere on the line between them. This is called coincidence detection, and it gives a direction without the heavy collimators that older cameras needed.
In the early 1950s, Gordon Brownell and William Sweet at Massachusetts General Hospital built positron-detecting devices for locating brain tumors, using coincidence between two detectors. These were simple scanners and did not make tomographic images, but they showed the principle. Detector crystals from the wider field of radiation detection improved steadily, and computer reconstruction methods developed for CT in the early 1970s were then applied to positron data on electronic computers.
The first PET scanners
In the early to mid-1970s, groups in the United States built the first tomographic positron scanners. Michael Phelps and Edward Hoffman, working with Michel Ter-Pogossian at Washington University in St. Louis, developed a scanner often cited as the first, sometimes called PETT (positron emission transaxial tomograph). Brownell’s group in Boston, and others in Europe, built competing designs. Credit for the first PET scanner is therefore shared, and accounts vary depending on how “first” is defined.
The tracer problem
A scanner alone is not useful without a suitable tracer. The isotopes used in PET, such as carbon-11, nitrogen-13, oxygen-15 and fluorine-18, have short half-lives, from about two minutes for oxygen-15 to nearly two hours for fluorine-18. This means they must be made close to where they are used, usually in a cyclotron, and attached to useful molecules quickly.
The breakthrough tracer was fluorodeoxyglucose (FDG), a glucose-like molecule labeled with fluorine-18. Chemists at Brookhaven National Laboratory and collaborators at the University of Pennsylvania developed it in the mid-1970s, and early human studies of brain glucose use followed. Because tissues with high energy demand take up more glucose, FDG-PET could show regional differences in brain and tumor activity.
Growth and combined scanners
PET stayed mostly in research centers through the 1980s because cyclotrons and radiochemistry were expensive. Over the 1990s it entered routine hospital use, especially in oncology, cardiology and neurology. Advances in detector crystals and electronics improved resolution and shortened scan times. Around 2000, David Townsend and Ronald Nutt introduced a combined PET/CT scanner, which lets doctors line up the metabolic image with an anatomical one. PET/MRI systems appeared later.
Cautions
PET involves ionizing radiation from the tracer, and CT in a combined scanner adds more. Images need careful interpretation, since normal organs and inflammation can also accumulate tracer. Whether a scan is appropriate for any person is a medical decision, not something this history can address.
What it added
PET’s contribution was to image biochemistry in the living body, including brain activity, without surgery. It joined X-ray, ultrasound, CT and MRI as another distinct window, and it is a good example of an invention that depended on many fields advancing at once.