ARRS Announces 2019 Scholars

The ARRS Scholarship supports study that will enable the scholar to attain professional career goals with the possibility of changing the way radiology is practiced. By giving both recognition and financial support to the activities and studies of young radiologists, ARRS helps these emerging leaders to prepare for positions of leadership in academic radiology. Nominees are submitted by medical schools, affiliated hospitals, and clinical research institutions with interests in training and research in diagnostic radiology, nuclear medicine, the basic sciences, or professions fundamental to imaging techniques.

Florian J. Fintelmann

Advancing Lung Cancer Care With Imaging Biomarkers

Lung cancer is one of the most frequently occurring and lethal types of cancer. As new therapeutic options are becoming available for lung cancer, innovative quantitative imaging biomarkers are required to support therapy selection, treatment monitoring, and drug development. With the support of the ARRS Scholarship, Florian J. Fintelmann aims to be at the leading edge to develop these tools by discovering and validating imaging biomarkers for prognostication in lung cancer patients while deepening the understanding of key molecular oncology concepts in lung cancer. 

Fintelmann’s project will expand the role of CT in lung cancer care beyond lesion detection, tumor staging, and surgical planning to patient-level prognostication. He hypothesizes that body composition metrics on chest CT reflect cardiopulmonary fitness and can improve overall survival estimates in patients with advanced lung cancer. Although the definition of sarcopenia has been well established at the lumbar spine, no reference values exist for the thoracic muscles and the relationship of sarcopenia to cardiopulmonary function and frailty remain to be defined. With the encouraging preliminary data he has generated, Fintelmann aims to better the understanding of the impact of body composition biomarkers on survival, which would inform surgical decision-making and facilitate patient-centered and individualized lung cancer care. He will determine the relationship between CT body composition metrics, frailty, and cardiopulmonary function; establish reference values for chest CT body composition metrics to support sarcopenia diagnosis on chest CT; and study the effect of body composition on overall survival in patient cohorts with advanced lung cancer who are receiving mutationspecific therapies or immunotherapy. 

Fintelmann is assistant professor of radiology at Harvard Medical School and staff radiologist at Massachusetts General Hospital. Prior to his appointment in the division of thoracic imaging and intervention, he completed radiology residency and fellowship training at Massachusetts General Hospital. Fintelmann graduated from medical school at the University of Ulm, Germany. 

Albert Hsiao

Adaptive Artificial Intelligence for the Acquisition and Analysis of Multiplanar MRI

Machine learning—and specifically, deep neural networks—promise to transform the practice of diagnostic radiology. These technologies continue to evolve at a rapid pace due to tremendous innovations in computational hardware and novel neural network architectures. Convolutional neural networks (CNNs) comprise an important subset of deep neural networks that have been adapted over the past several years to address problems in computer vision. Although the technical performance of these new algorithms for computer vision are impressive, clinical usage of this technology presents several complexities that are new and unfamiliar to the machine learning community. With the support of the ARRS Scholarship, Albert Hsiao aims to develop a framework for developing, deploying, and adaptively training these networks to realize their potential clinical value. 

Hsiao’s project will evaluate a prototype system of CNNs that his team has developed to automate scan prescription of multiplanar cardiac MRI through automated localization of cardiac structures. This technology has the potential to reduce the training and expertise required for a technologist to perform cardiac MRI and increase accessibility to this examination beyond a limited set of high-end clinical centers. He will also develop a system to automate the process of “iterative refinement” that is currently undertaken manually, with the goal to continually improve performance of the neural networks that are used. This work will lay a foundation for an adaptive system of neural networks to improve performance of the diagnostic modalities where they are employed, thus pioneering methods to make this technology adaptive and robust for everyday clinical practice. 

Albert Hsiao is assistant professor of radiology at the University of California, San Diego (UCSD), where he founded the Augmented Imaging/Artificial Intelligence Data Analytics laboratory to further build advanced imaging techniques for earlier, more precise diagnosis and treatment in cardiovascular disease and oncology. After completing a dual major in biology and engineering/computer science at the California Institute of Technology, he attended medical school at UCSD as part of the dual MD-PhD Medical Scientist Training Program. He is a graduate of the UCSD bioengineering and bioinformatics PhD programs, completing his thesis work on the functional genomics of insulin resistance.

Hsiao continued his clinical training at Stanford University in general surgery, diagnostic radiology, interventional radiology, and cardiovascular imaging. As a clinical diagnostic radiology resident, he cofounded Arterys, Inc., a start-up company that developed a cloud-based medical imaging artificial intelligence platform and partnered with GE Healthcare to bring 4D flow MRI technology to market.

Richard Duszak, Jr.

The Radiology Malpractice and Risk Management Actionable Knowledge Advancement and Dissemination Initiative

Malpractice litigation can impact how healthcare services are delivered across the United States, with malpractice concerns resulting in unnecessary medical imaging and higher healthcare costs. Historically, malpractice research has focused on individual case studies published in single case reports, which are devoid of contextual frequency at the population level. All the while, malpractice education has largely relied on traditional in-person lectures, which limit the ability to disseminate information broadly and rapidly. With the support of the ARRS Leonard Berlin Scholarship, Richard Duszak aims to fill these gaps in scholarship and teaching by developing a deeper and broader understanding of the contributors to and drivers of malpractice litigation and better disseminating such knowledge to reduce radiologists’ collective malpractice exposure and improve the quality and value of radiology services to patients. 

Duszak’s project will advance the depth and breadth of radiology malpractice research in a multidisciplinary manner. He will develop scalable and actionable approaches to using legal data sets to more meaningfully identify and characterize radiology malpractice case cohorts. He will also advance the specialty’s traditional approaches to malpractice knowledge dissemination by creating an accessible and robust electronic learning toolkit to provide trainees and practicing radiologists alike with practical and actionable education in radiology malpractice and risk management issues. These combined and complementary scholarship and educational initiatives will together stimulate, advance, promote, and disseminate radiology malpractice knowledge throughout the radiology community. The parallel research and teaching programs will, through increasing awareness, help radiologists mitigate their tort exposure by reducing medical errors and improving communication with patients and referring physicians alike. The combined result will be improved patient care. 

Duszak is professor of radiology and vice chair for health policy and practice at Emory University’s School of Medicine. Nationally recognized for his work in imaging health policy, Duszak came to Emory after previously serving as founding CEO of the Harvey L. Neiman Health Policy Institute and president of a regional health system–based radiology practice. Long active in organized medicine and physician payment system development, Duszak is keenly interested in studying the intersection of health care delivery and payment systems to identify new opportunities for creating value and sustainability in fragile health care markets and sectors, particularly as they relate to medical imaging. His scholarly work to date has largely focused on medical imaging utilization, delivery systems, and emerging value- and quality-focused payment methodologies. 

More from