18 novembre 2006 6 18 /11 /novembre /2006 12:02


NMR Imaging and Its Application to Clinical Medicine

The idea of using a magnetic gradient to introduce spatial information into signals from an NMR spectrometer, which can then be converted to an actual visual image, was Paul Lauterbur's. The idea of applying NMR to cancer detection and realizing the idea in the first commercial MRI machine were Raymond Damadian's contribution. [50]

Raymond Damadian received his M.D. degree from the Albert Einstein College of Medicine at Yeshiva University, and did his internship and residency at the Downstate Medical Center in Brooklyn, part of the State University of New York system. [51] He chose to pursue a research career rather than practice clinical medicine. He joined SUNY in 1967 with a joint appointment in Biophysics and Internal Medicine. During the years before the work that led directly to publication of his first Science article, Damadian was working on cellular composition and chemical transport. This work, focusing on cell metabolism, led Damadian to believe that there should be a way to detect cancerous cells through chemical analysis rather than by relying on direct visualization under a microscope. References in a standard chemistry textbook to Nicolaas Bloembergen's finding, 20 years earlier, that NMR relaxation times T1 and T2 were affected by the viscosity of a fluid led Damadian eventually to consider applying NMR spectroscopy to tissue in the hope of finding differences between cancerous and normal tissue.

At the April 1969 meeting of the Federation of American Societies in Experimental Biology, Damadian and Freeman Cope agreed to conduct NMR experiments on detecting potassium in bacteria from the Dead Sea. (Cope had been working on detecting sodium in brain tissue and wanted to measure potassium in biological tissue.) They were able to borrow time on a new, pulsed FT spectrometer made by NMR Specialties of New Kensington, PA, which enabled them to observe relaxation times directly. They were successful in detecting potassium, and this led Damadian to seek support from New York City's Health Research Council for purchase of a pulsed NMR spectrometer to explore the potential use of spectroscopy "for early non-destructive detection of internal malignancies" (letter quoted in Mattson and Simon, 1996: 646). A year later, Damadian had assembled a collection of rats bearing tumors and had again received permission to use spectrometers at NMR Specialties. The T1 measurements Damadian made of cancerous vs. normal rat tissue were the basis for his 1971 article in Science.

Subsequent investigations by Damadian and others revealed that, although the relaxation times of signals from cancerous tissue were different from those from normal tissue, they overlapped with relaxation times from noncancerous but abnormal tissue. But Damadian was initially convinced that relaxation times could be used to detect cancer, and therefore in 1972 filed a patent claim for an "Apparatus and Method for Detecting Cancer in Tissue." The patent included the idea of using NMR to "scan" the human body to locate cancerous tissue. In early 1976, he and his graduate students began working on a prototype machine, with technical assistance from the physics department at Brookhaven National Laboratory (which put him in touch with people designing superconducting magnets for the latest particle accelerator) and financial assistance from private donors. Work continued for a year and a half on an NMR machine with a superconducting magnet large enough to accommodate a human body. In July 1977, Damadian and his students succeeded in creating a crude image of the cross section of a human chest, accomplished by moving the subject through 106 slightly different positions to build up an image. (The focal point was achieved through a combination of electrically focusing the RF field and taking advantage of inhomogeneities in the magnet's main field.)

Nine months after achieving the chest image, Damadian left the university to set up FONAR Corporation. He and a staff of about 30, working in a rented facility, produced a permanent-magnet MRI machine and introduced it at the 1980 meeting of the American Roentgen Ray Society and, later that year, at the annual meeting of the Radiological Society of North America. Subsequent models of the FONAR machine adopted Lauterbur's projection method, but that approach was quickly superseded by the pulsed Fourier transform method that became the dominant design of the MRI industry.

Paul Lauterbur received a B.S. in chemistry from Case Institute in 1951. He took a job with the Mellon Institute in Pittsburgh and, after serving in the Army, returned to Mellon in 1955. While in the Army, he helped set up an NMR laboratory and began research on NMR spectroscopy. He then set up his own NMR lab at Mellon and produced numerous papers. He worked on his Ph.D. in chemistry at the University of Pittsburgh while at Mellon, receiving the degree in 1962. He moved to SUNY Stony Brook's chemistry department in 1963 and continued his NMR studies, focusing on intramolecular and solvent isotope effects on chemical shifts. Lauterbur says that two areas that began to interest him at Stony Brook led toward imaging. One was biological applications of NMR, which led him to do a sabbatical in 1969-70 at the Stanford Medical Center, where he worked on labeling of proteins and tritium NMR. The second was computer-aided acquisition and processing of NMR spectra. Because the Stanford experiments were not productive, Lauterbur had difficulty obtaining support for his subsequent research (most proposals were submitted to NIH).

Through an unusual confluence of events, Lauterbur ended up at NMR Specialties in the spring of 1971, observing researchers from Johns Hopkins attempt to replicate Damadian's results. While at Mellon, Lauterbur had helped to set up NMR Specialties and, as a result, was on its Board of Directors in 1971. Because the company was about to go bankrupt, Lauterbur agreed to take over the company (a decision made easier because he did not have summer salary or grant support that year). At NMR Specialties that summer, Lauterbur watched as Leon Saryan from Johns Hopkins used the same machine Damadian had used to compare the relaxation times of fast- and slow-growing tumors in rats with those of normal rat tissue. Immediately afterward, Lauterbur went out for a hamburger, thinking about how the information obtained from invasive techniques might be obtained in other ways, so that the location of a signal within an object might be identified. That evening, he struck on the idea that inhomogeneous magnetic fields introduced locational coordinates into NMR signals, and immediately bought a notebook, wrote the ideas down, and had them witnessed.

During the next several days, Lauterbur figured out how to create a series of one-dimensional projections by changing the orientation of the gradient field and then mathematically reconstruct a two-dimensional image. (Lauterbur was not aware of previous techniques for accomplishing this, nor did he know that the principle was being applied to CT scanning at about this time.) He attempted to patent the idea privately but failed, and subsequently (when back at Stony Brook) could not arouse the interest of Research Corporation, which handled intellectual property rights at Stony Brook (SRI interview, April 12, 1996). He turned to experiments at Stony Brook in 1972, supported by part of an institutional research grant to Stony Brook from NIH. Lauterbur used a pair of capillary tubes filled with water and D2O in one experiment and water and a solution of MnSO4 in water in a second. The results were fuzzy but recognizable pairs of images of the cross-sections of the tubes, images that changed as a result of the effect of different RF power levels and T1 relaxation times of the different liquids. (His initial results were obtained by hand on graph paper rather than on a display tube. A typewriter typed out the 400 spin density figures in a 20 20 matrix.) The paper he submitted to Nature describing the experiment was published in 1973 (Lauterbur, 1973). The paper explicitly mentions the possible application of NMR imaging to the "in vivo study of malignant tumors, which have been shown to give proton nuclear magnetic resonance signals with much longer water spin-lattice relaxation times than those in the corresponding normal tissues." At about this time, Lauterbur obtained a grant from the National Cancer Institute that enabled him to buy an NMR spectrometer and develop a research team (Mattson and Simon, 1996: 722).

After his initial publication, Lauterbur gave numerous talks in universities and industry, but aroused little interest in the latter. He visited the United Kingdom and interacted with the Nottingham physicists, specifically Hinshaw, Moore, Andrew, and Mansfield. He also visited Hounsfield at EMI. He traveled to the United Kingdom frequently because he received more positive reactions from the British researchers, who were convinced something important would come from their work. He consulted with industry and was visited in his lab by industry. No one treated his work seriously from a commercial point of view; he tried to sell his ideas at Varian but was unsuccessful. Technicare came to his lab at Stony Brook, but nothing came of it. As Chen and Hoult point out,

Curiously, with the exception of the EMI company in Britain, ... industry took little or no notice of Lauterbur's invention, and it was left to universities to develop magnetic resonance imaging. (Chen and Hoult, 1989: 40)

During the next 10 years, Lauterbur and his students and colleagues conducted studies of the chemical shift imaging of proton signals, the introduction of paramagnetic contrast agents, and three-dimensional projection reconstruction and, toward the end of this period, a whole-body 3D system in which cancer specimens that had been removed through surgery was studied (Mattson and Simon, 1996: 727). In 1985, he left Stony Brook to join the faculty of the University of Illinois at Urbana-Champaign, where he is now Director of the Biomedical Magnetic Resonance Laboratory.

NSF Role. Our interview with Lauterbur revealed that, for most of the period during which he was building the basis for his fundamental contribution, he struggled to obtain support, mostly from NIH. As we saw, at the time of his initial experiments, he was working with a small amount of money from Stony Brook that in turn came from an NIH institutional grant. His first interaction with NSF came in the 1980s, when he wanted to get started on nonbiological materials. He did obtain support from the RANN (Research Applied to National Needs) Program, but the program was terminated in the middle of his grant period. Our search of the NSF awards database showed that Lauterbur received a 2-year award in 1966 for $61,000, "Anisotropic Nuclear Magnetic Shieldings in Solids," and another in 1980 for $98,000 for 30 months, "Nuclear Magnetic Resonance Microscopy." A search of the database using the keywords "magnetic resonance imaging" yielded just two awards during this period: $309,000 to Jerome Singer[52], an electrical engineer at the University of California-Berkeley, in 1977 for "In-Vivo Nuclear Magnetic Resonance Imaging of Flow Patterns" and $108,000 to Tara Das at SUNY Stony Brook in 1987 for "Theory of Relaxation Times Associated with Contrast Agents in Magnetic Resonance Imaging." A search of the database using "magnetic resonance" yielded more than 900 small research awards from 1955 to the present, totaling more than $90 million. A large proportion of these awards were for the purchase of NMR spectrometers. The NIH awards database, which is accessible on-line, shows that Lauterbur was awarded his first NIH grant in 1972, and from then on averaged about two a year through 1984. In 1974, he received $232,000 from NIH for "Application of NMR Zeugmatography in Cancer Research." [53]

At SUNY, following publication of his NRM experiments with rat tumors, Damadian received grants from the American Cancer Society, the New York City Health Research Council, and private philanthropists. Long Island investors provided the capital to help start FONAR (SRI interview with Damadian, August 17, 1996). A search of the NSF database confirmed that Damadian has not received support from NSF.

centrale nucléaire .Gravelines , fort-philippe .



Partager cet article

Repost 0