Tribute

Dr. Robert F. Wagner was a distinguished research physicist and member of the Senior Biomedical Research Service in the Center for Devices and Radiological Health (CDRH), U.S. Food and Drug Administration (FDA). His career was dedicated to the development of consensus measurement methods for the assessment of medical imaging systems, quantitative medical imaging and tissue characterization, and computer-aided diagnosis. He earned an international reputation in these areas and applied his expertise to a wide range of regulatory issues central to FDA’s mission. He enlightened the scientific community within the Agency as well as the international scientific community through the many invited presentations and tutorials he gave in and outside FDA, his numerous publications, his many professional society activities, and his assistance in regulatory decision making. His greatest legacy may be the many young scientists he nurtured, either working directly under his tutelage at the FDA or otherwise benefiting from his unfailing patience and unselfish ease of availability. Robert F. Wagner was born in Philadelphia on January 10, 1938. After graduate and post-graduate work on the physics of nuclear interactions with radiation, he was hired by the Bureau of Radiological Health (a precursor to CDRH) to assess the dose reduction potential of radiographic intensifying screens made with phosphors developed in the color TV industry. Within three months he published a review of the relevant imaging literature from the medical, defense, consumer, and scientific communities, together with a charter for a laboratory program. Soon thereafter, Bob introduced digital noise analysis to radiography, and showed that the new technology offered a 1.6 to 2.5-fold exposure reduction without compromising imaging performance. He then launched a program of inter-laboratory comparison of measurements on radiographic film samples that were circulated among fifteen commercial, government, and academic laboratories world wide. In the process he became the prime mover for work toward consensus methodology for quantitative imaging performance measurements.

In the mid-1970s the new technology of computed tomography (CT) entered the marketplace and permanently revolutionized medical imaging. However, most investigators were concerned by the high exposures and doses required. The mathematical community began to promise exposure reductions through the development of “low-dose” computer or image processing algorithms. Wagner and his colleagues showed that CT was intrinsically a high-dose technique -- the concept of a low-dose algorithm was a fiction – and in fact CT systems could be shown to be highly efficient in their use of the detected radiation, although there were geometrical or collimation problems (remedied by later CT generations). The understanding achieved by Wagner and his coworkers during this period was used by the BRH laboratory staff when the CT amendments to the X-ray Performance Standards were written.

Bob Wagner subsequently lead the development of a “unified signal-to-noise” ratio approach to the assessment of all conventional modes of medical imaging. Using principles of statistical decision theory, he showed how the transfer functions and noise spectra of imaging systems could be used to predict the detectability and discriminability of lesions in all the common medical imaging modalities. In particular, this work lead to the wide use of detection quantum efficiency as a performance metric for radiographic imaging systems that continues to this day.

Dr. Wagner’s ability to unify performance measures among different imaging modalities made him especially valuable to the Agency. However, medical ultrasound initially presented a particularly difficult challenge to Dr. Wagner’s approach because of its very singular noise (speckle) characteristics. Here again he was able to bring to bear key results from other fields, notably the radar and laser literature, to develop the first satisfactory theoretical and experimental approach to quantitative understanding of the speckle, its dependence on ultrasonic hardware parameters, and its effect on the detectability of lesions by medical ultrasound. Two back-to-back papers on these issues were published by the IEEE and were given the “Best paper Award for 1983” by the Institute’s Sonics and Ultrasonics Division. These and the stream of related publications on ultrasound that followed made possible many advances in the development of standards for ultrasonic imaging system performance assessment.

Having achieved a quantitative understanding of ultrasound speckle, Dr. Wagner initiated a sophisticated investigation of the information characterizing the tissue masked behind the speckle yet potentially separable from it. He made highly original contributions to the field of quantitative ultrasonic tissue characterization by showing that the incoherent backscattered signal could be separated from two classes of coherent ultrasound signals returned from the body. He and his collaborators subsequently carried out a study to demonstrate the clinical potential of a resulting algorithm (which was awarded patent protection) for detection and discrimination of several diffuse liver diseases from each other and from normal liver. The resulting paper was published in Investigative Radiology, the journal of the Association of University Radiologists, in 1989 and received the Stauffer Award for the best clinical paper in that journal.

Of course, for most imaging systems, the final measure of performance is the ability to convey information to a human observer: the clinician. Dr. Wagner and his collaborators performed key investigations on the ability of human observers to extract information from images obtained by a medical imaging system. These studies demonstrated that there are circumstances where the human observer is extremely efficient, so that the information in the image is readily extracted. They also showed that there are regimes where the efficiency of the human is markedly reduced, indicating imaging modalities that would benefit greatly from adjustments in acquisition hardware or preprocessing algorithms to render the information more readily extractable by the clinician. Alternatively, such systems are fertile ground for the use of computer-assisted diagnosis strategies: indeed, it was this recognition, combined with the necessary statistical and physical tools, that lead to the project on quantitative tissue characterization of abdominal organ tissue via ultrasound described above. The earliest work by Dr. Wagner and his colleagues on human visual detection efficiency resulted in a Science paper in 1981.

In recent years, Dr. Wagner extended his investigations of human performance to explore statistical methods for analyzing the performance of imaging systems within the context of reader variability (differences in reader skill and decision criteria). In addition, Dr. Wagner had been working to extend his image assessment methodology to the many new and increasingly sophisticated computer techniques being developed to aid the human reader in the interpretation of high-dimensional image data sets for medical diagnosis. In particular, he was working on study designs, objective measurements, and analytical methods for the assessment of stand-alone diagnostic modalities such as high-dimensional DNA micro-arrays.

Dr. Wagner’s research resulted in a stream of highly cited and extremely creative scientific publications, as well as recognition in the form of honors and awards by FDA and his professional community. He was elected to the rank of Fellow by five societies: IEEE, AIMBE, OSA, SPIE, and SPSE. Within the FDA he was awarded the FDA Commendable Service Award and the Award of Merit, the Commissioner’s Special Citation, and the Public Health Service Superior Service award, in addition to having been a member of numerous groups receiving unit awards.

In recognition of his leadership in the field of assessment of diagnostic imaging performance, Dr. Wagner was chosen as a principal author of an International Commission for Radiation Units and Measurements (ICRU) report on image quality in medical imaging. The resulting document was published by the ICRU during the centenary year (1995) of the discovery of x rays by Roentgen. This major document laid the foundation for a series of ICRU reports with more detailed recipes, one medical imaging modality at a time, that have been developed since

Dr. Wagner served on numerous academic advisory boards, search committees, conference program committees, and editorial boards, and carried on an extremely active service function as a reviewer of grants and scientific papers. His very active grant review activities benefited such institutions as the National Science Foundation, the National Cancer Institute, the Atomic Energy Commission, the Canadian Research Councils and those of Great Britain, and the National Institute of Dental Research. He was a very prolific reviewer, reviewing papers for a broad spectrum of journals, including Medical Physics, Physics in Medicine & Biology, Optical Engineering, the Journal of the Acoustical Society of America, and many others. In 2001 he co-chaired the annual conference of the Medical Imaging Perception Society.

Dr. Wagner revolutionized the way in which medical imaging technologies are evaluated. He spent more than 35 years in seminal work on understanding and developing paradigms for the description and evaluation of medical imaging and computer-assisted diagnostic modalities. His contributions to the science enterprise will live on in the work of his students and colleagues, and through his publications, for decades to come.

In addition, Dr. Wagner had a strong passion for science and humanity, a warm and charming personality, a wonderful wit, and a self-deprecating humility. He will be greatly missed.

Education

1959 B.S., Electrical Engineering, Villanova University, PA

University Valedictorian

Selected "Outstanding Graduate" by American Association

Of University Professors at Villanova University

1965 M.A., Theology, Augustinian College, Washington, D.C.

1965 M.S., Physics, Catholic University, Washington, D.C.

1969 Ph.D., Physics, Catholic University, Washington, D.C.

Professional Experience

1970 - 1972 Post-Doctoral Research Associate, Physics Department,

Ohio University, Athens, Ohio

1972 - 1976 Research Physicist, Division of Electronic Products, BRH

1976 - 1994 Chief, Diagnostic Imaging Section, BRH/CDRH/FDA

1995 - present FDA Senior Biomedical Research Scientist (SBRS)

Selected Honors and Awards

Fellow – AIMBE - American Institute for Medical and Biological Engineering (2001).

Charter Member of IEEE Medical Imaging Society (1996).

Fellow - IEEE - The Institute of Electrical and Electronics Engineers (1991).

Fellow - OSA - The Optical Society of America (1989).

Fellow - SPIE - The International Society for Optical Engineering (1988).

Fellow - SPSE - The Society for Imaging Science and Technology (1983).

Alumni Award for Professional Achievement: Villanova University (1989).

Best Paper Awards: IEEE Ultrasonics (“Statistics of Speckle in Ultrasound B-Scans”) 1983; Investigative Radiology, 1989, Garra et al.

Kyle Myers