First Positron Imaging Device - 1950

The first application of positron annihilation radiation for medical imaging is well documented. In a discussion with William Sweet, then the Chief of the Neurosurgical Service at the Massachusetts General Hospital (MGH), in the early part of 1950, I made several suggestions to improve the quality of nuclear images for the detection of brain tumors and other brain diseases. In particular, I suggested that the use of annihilation radiation following positron emission might improve the quality of brain images by increasing sensitivity and resolution. The Physics Research Laboratory (PRL) at MGH had just been established under my direction and, with support from the Neurosurgical Service, a simple positron scanner using two opposed sodium iodide detectors was designed and built within six months. Imaging of patients with suspected brain tumors was commenced almost immediately. The results were sufficiently encouraging that an addendum including results on positron imaging was included in a paper by Sweet on brain tumor localization. The paper was then in press in the New England Journal of Medicine and together with the addendum appeared in December of 1951. During the same year, a paper by Wrenn, Good and Handler described independent studies on annihilation radiation detection. These authors went on to illustrious careers - Philip Handler became the President of the American Academy of Science - but did not publish further on this topic.

Despite the relatively crude nature of this imaging instrument, the brain images were markedly better than those obtained by other imaging devices. It also contained several features that were incorporated into future positron imaging devices. Data were obtained by translation of two opposed detectors using coincidence detection with mechanical motion in two dimensions and a printing mechanism to form a two-dimensional image of the positron source. This was our first attempt to record three-dimensional data in positron detection. An article published in 1953 described this device and included preliminary results (Brownell and Sweet 1953).