Critical to making MRI a reality was the advent of the high speed computers needed to handle the enormous quantity and complexity of the computations involved in imaging. In addition to the necessary computing power, three other developments contributed to the birth of MRI. One was the work of British electronics engineer Godfrey Hounsfield, who in 1971 built an instrument that combined an x-ray machine and a computer and used certain principles of algebraic reconstruction to scan the body from many directions--manipulating the images to produce a kind of cutaway view of the interior. Unknown to Hounsfield, South African nuclear physicist Allan Cormack had published essentially the same idea in 1963, using a reconstruction technique called the Radon transform. Although Cormack's work was not widely circulated, in 1979 he and Hounsfield shared the Nobel Prize in physiology or medicine for the development of computerized tomography, or CT. The principles underlying CT are the foundation of many sophisticated imaging methods in use today.
The other two developments essential to MRI were related to nuclear magnetic resonance
. One was the conceptualization of NMR as a medical diagnostic tool; the other was the invention of a practical method for producing useful images from NMR data.
As early as 1959, J. R. Singer at the University of California, Berkeley, proposed that NMR could be used as a non-invasive tool to measure in vivo blood flow. Then in 1969, Raymond Damadian, a physician at Downstate Medical Center in Brooklyn, New York, began to think of a way to use the technique to probe the body for early signs of cancer. In a 1970 experiment he surgically removed fast-growing tumors that had been implanted in lab rats and showed that the tumors' NMR signals differed from those of normal tissue. Damadian published the results of his experiments in 1971 in the journal Science.
An essential technical advance that opened up the ensuing widespread application of NMR to produce useful images was due to chemist Paul Lauterbur, who was then at the State University of New York at Stony Brook. In 1971, he watched a chemist named Leon Saryan repeat Damadian's experiments with tumors and healthy tissues from rats. Lauterbur concluded that the technique was insufficiently informative for locating and diagnosing tumors and went on to devise a practical way to use NMR to make images.
Lauterbur's groundbreaking idea was to superimpose on the spatially uniform static magnetic field a second weaker magnetic field that varied with position in a controlled fashion, creating what is know as a magnetic field gradient. At one end of a sample the graduated magnetic field would be strong, becoming weaker in a precisely calibrated way down to the other end. Because the resonance frequency of nuclei in an external magnetic field is proportional to the strength of the field, different parts of the sample would have different resonance frequencies. Thus, a given resonance frequency could be associated with a given position. Moreover, the strength of the resonance signal at each frequency would indicate the relative size of volumes containing nuclei at different frequencies and thus at the corresponding position. Subtle variations in the signals could then be used to map the positions of the molecules and construct an image. (Today's magnetic resonance imaging devices impose three sets of electromagnetic gradient coils on the subject to encode the three spatial coordinates of the signals.)
Across the Atlantic in Britain, Peter Mansfield at the University of Nottingham, England, had a similar idea. He was looking into using NMR to obtain structural details of crystalline materials. In work published in 1973, Mansfield and his colleagues also used a field gradient scheme. In 1976, Mansfield developed a MRI technique known as echo-planar imaging, which can rapidly scan a whole brain.
Meanwhile, Lauterbur's results, published in 1973, included an image of his test sample: a pair of small glass tubes immersed in a vial of water. Working with the small NMR scanner he had created (and using a technique called back projection borrowed from CT scanning), he continued to image small objects, including a tiny crab scavenged by his daughter from the Long Island beach near his home. By 1974, using a larger NMR device, he produced an image of the thoracic cavity of a living mouse. Mansfield, for his part, had imaged a number of plant stems and a dead turkey leg by 1975, and by the next year he had captured the first human NMR image--a finger. Damadian also was at work producing images. In 1977, he produced an image of the chest cavity of a live man.
By the early 1980s the flurry of activity around MRI had given rise to a burgeoning commercial enterprise. ("Nuclear" had been quietly dropped from the name in the meantime because of its unfavorable connotations.) Advances in high-speed computing and superconductive magnets allowed researchers to build larger MRI machines with enormously improved sensitivity and resolution and made possible many new applications.
