Dr. Ehsan Samei, PhD, is a Professor of Radiology, Medical Physics, Biomedical Engineering, Physics, and Electrical and Computer Engineering at Duke University, where he also serves as the director of Carl E. Ravin Advanced Imaging Laboratories and the founding director of the Clinical Imaging Physics Group. His expertise includes x-ray imaging, theoretical imaging models, simulation methods, and experimental techniques in medical image formation, analysis, assessment, display, and perception. His current research includes methods to develop image quality and dose metrics that are clinically relevant and that can be used for optimum design and utilization of advanced imaging techniques aimed to achieve optimum interpretive, quantitative, and molecular performance. Five areas of recent activities are outlined below.
1. Objective assessment and clinical relevance of image quality:
While image quality is a common term in radiology circles, its quantification has proven to be complex, as many factors contribute to the overall degradation of a medical image. Dr. Samei"s research and prior publications have provided a framework for assessing the performance of digital radiography systems, as recently reflected in an international standard on image quality measurements (IEC 62220- He is currently pursuing methods to further streamline the assessment methodologies and include other important contributing yet hitherto ignored factors such as scattered radiation. Furthermore, the connection between diagnostic accuracy and the scientific metrics of image quality is not straightforward. Two studies are now in progress to substantiate the relevance of these quantities in radiography (chest and breast) and in computed tomography (pediatric CT). The former investigation, initiated in his laboratory, is supported by a pre-doctoral grant from the DOD. The pediatric CT research initiative, supported by a current industrial grant from GE, aims to investigate strategies to reduce radiation dose in pediatric CT without compromising diagnostic quality.
2. Quantitative imaging:
By and large, since its inception, radiology as a discipline has been developed as a qualitative discipline in which physicians "interpret" medical images to gain diagnostic insights. However, current, mostly digital, medical images contain a large amount of information, which if effectively harnessed, can be invaluable in the quantification of the disease, in the assessment of the effectiveness of various therapeutic regimens, and in providing tractability of the medical information toward evidenced-based and patient-specific medicine. The main technical obstacle preventing this potential has been the lack of certainty about the extent to which image data can be analyzed quantitatively in light of various sources of variability (eg, case variability, patient positioning, system variability, etc). However, there is little doubt that the future of radiology is quantitative, and additional research is warranted to lead to that transition.
Dr. Samei has recently initiated a number of research projects aiming to assess the relative magnitude of the sources of variability in diagnostic radiology, and from that to devise approaches by which robust quantitative data can be extracted from medical images. Initial projects have focused on CT and breast tomosynthesis. Additional projects are planned for digital mammography and digital radiography. He is also fostering additional quantitative imaging projects led by other faculty members in RAI Labs. With the preliminary results of these ongoing initiatives, he aims to lead RAI Labs toward a comprehensive program project on quantitative imaging.
X-ray imaging is now used extensively throughout the world due to its low cost, widespread availability, speed, ease of interpretation, and exquisite representation of anatomy. However, it has to date been limited to structural imaging. But, the underlying processes behind many human diseases occur at the molecular level, suggesting many advantages that the imaging of these molecular and functional processes can provide. In spite of the dominance, cost-effectiveness and maturity of x-ray imaging, almost all molecular imaging investigators have focused their research on non-x-ray based technologies such as MRI, nuclear medicine, and optical imaging techniques.
Dr. Samei has recently initiated a few preliminary studies to assess how conventional x-ray imaging can be extended to provide molecular and functional information. The overall goal of his investigation is to develop x-ray-based molecular and functional imaging methods. While maintaining the key advantages of x-ray imaging, the proposed method will provide functional images that have higher spatial and temporal resolution than what is currently possible through more "conventional" functional imaging methods (eg, Nuclear Medicine and fMRI), and that are inherently co-registered to anatomical information. The method employs nano-particle, "smart," liposomal contrast agents to reveal specific functional and molecular processes and physiological functions within the human body, initially for the purpose of early detection of cancer and cardiovascular disease, the two leading causes of death world-wide. The pursuit of this project is particularly timely in light of recent developments in the fabrication of stable nano-particle liposomes and the development of antibody labeling methods. The application of functional and molecular imaging to x-ray technology has the potential to profoundly impact global clinical practice by spreading the benefits of functional imaging through a more accessible technology, unleashing the power and the promise of molecular imaging for many.
4. Correlation and stereo imaging for improved early detection of cancer:
The early detection of lung cancer has been one of the outstanding challenges in radiographic imaging, the significance of which can be discerned only by considering the fact that lung cancer remains the leading cause of cancer death in the US, surpassing breast, prostate, colon, and cervical cancers combined. His prior research has shown that interference of the anatomical structure is the dominant factor in the low detection of early lung cancer in radiographic images. Grounded on this basic understanding, Correlation Imaging (CI) aims to develop a more sensitive image acquisition and processing approach that minimizes this influence, and therefore improves the early detection of lung cancer. The NIH awarded Dr. Samei R21 and R01 grants to study the feasibility of this imaging approach. With that funding, chest correlation and stereo imaging was developed from an initial design into a state-of-the-art prototype imaging system (fully developed and integrated at Duke). The prototype equipment is now being used in a clinical trial to assess the clinical utility of this novel imaging technique. As of now, more than 80 patients have been imaged, and the preliminary findings are very encouraging. Furthermore, Dr. Samei significantly contributed to and served as a key investigator on an R01 grant aimed to develop a cone beam CT imaging system (PI: Martin Tornai, PhD) and tomosynthesis (PI: Joseph Lo, PhD) for breast imaging at Duke. Dr. Samei has an active interest in developing other advanced x-ray imaging techniques for the early detection of cancer including limited angle and inverse geometry CT.
5. Objective assessment and impact of display quality on diagnostic accuracy:
The way in which medical image data are displayed has a direct influence on diagnosis. This dependency is task-specific, and for many tasks in medical imaging, including the early detection and classification of cancer, has not been fully substantiated in quantitative terms. For the last few years, Dr. Samei has led a national task force (AAPM TG18), and served on an ACR and an IEC committee to define standard testing methodologies for medical display devices. His research in this area is currently focused on the influence of display characteristics on the diagnostic interpretation of breast cancer and lung cancer, the fine-tuning of new testing methodologies, and on the influence of ambient light in diagnostic accuracy. These studies are funded by the NIH and the display industry.