A total of 11 cutting-edge scientific studies will be presented at the 2024 SPIE Medical Imaging conference by researchers from the Center of Virtual Imaging Trials (CVIT) and RAILabs. SPIE Medical Imaging conference, this year co-chaired by our TRD3 leader Dr. Joseph Lo, is scheduled to take place from February 18-22 in San Diego, CA. Experts and researchers from CVIT and RAILabs will unveil their contributions across diverse research areas and modalities solidifying the commitment of these institutions to advancing the field of medical imaging. The array of presentations will encompass topics ranging from the clinical evaluation of automated exposure control systems in radiography to breakthroughs in photon counting CT technology. Additionally, attendees can anticipate insights from virtual imaging trial studies and the optimization of contrast media administration, showcasing the multifaceted expertise of these institutions. Finally, noteworthy is the presence of plenary speaker Cynthia Rudin from the Interpretable Machine Learning Lab at Duke, who will share insights into interpretable deep learning in medical imaging. Following is the list of the studies from CVIT and RAILabs: Monday 02/19 Francesco Ria: Performance assessment of photon counting versus energy integrated CT: concordance of in vivo and phantom measurements Amar Kavuri: Quantitative accuracy of lung function measurement using parametric response mapping: A virtual imaging study Xuan Liu: A residual-attention multimodal fusion network (ResAMF-Net) for detection and classification of breast cancer Wednesday 02/21 Fakrul Islam Tushar: Virtual NLST: towards replicating national lung screening trial Mridul Bhattarai: A generic model of semiconductor-based photon-counting detectors for spectral CT Cindy McCabe: A comparative study on the utility and choice parameters of photon-counting CT for bone quantification Darin Clark: Multi-Channel Reconstruction (MCR) Toolkit v2.0: open-source Python-based tools for X-ray CT reconstruction Nicholas Felice: Enhanced CT simulation using realistic vascular flow dynamics Thursday 02/22 Saman Sotoudeh-Paima: Development and application of a virtual imaging trial framework for longitudinal quantification of emphysema in CT Cindy McCabe: Synthesizing heterogeneous lung lesions for virtual imaging trials David Kim: Random walk small intestine models for virtual patient populations Stay tuned on our social media for a wealth of scientific knowledge, collaborative discussions, and pioneering discoveries at the SPIE Medical Imaging conference. For more details and to join this exciting journey, visit the official SPIE Medical Imaging website: https://spie.org/conferences-and-exhibitions/medical-imaging.
Dr. Mojtaba Zarei’s doctoral journey at Duke University was a captivating exploration of machine learning’s potential to revolutionize medical imaging, empowered by the innovative tool of virtual imaging trials. His research pursued two key objectives: optimizing scanner protocols for specific applications like cardiac and lung cancer screening, and developing AI models to enhance and harmonize image consistency. Additionally, he delved into the fascinating realm of material decomposition AI for spectral images and novel liver lesion detectability index. We asked Dr. Zarei to share some memories about his doctoral experience with us. Dr. Zarei, how the Center for Virtual Imaging Trials enhanced your Duke graduate student experience? Reflecting on my PhD experience, I am deeply grateful for the invaluable lessons I have gained. Participating in diverse group meetings, ranging from TRD2 with Dr. Abadi to TRD3 with Dr. Lo, honed my critical thinking skills and equipped me with valuable research development strategies. I mastered the art of managing research projects, navigating challenges with agility and adaptability, and independently taking ownership while seeking guidance when needed. Most importantly, I discovered the transformative power of collaboration, witnessing firsthand how diverse perspectives can synergistically create groundbreaking results. Additionally, I learned the crucial value of being a valuable and contributing team member. Your PhD path was strongly affected by the COVID-19 pandemic. How challenging was adapting to remote working and studying? Beyond the technical aspects, certain moments resonated deeply on a personal level. As an extrovert, adapting to the virtual interactions of joining Railabs during the early days of the COVID-19 pandemic presented a unique challenge. However, the subsequent return to in-person meetings and reconnecting with colleagues face-to-face was a truly special experience. While I did miss the casual camaraderie of lunches at the Railabs kitchen, the joy of reconnecting in person was truly profound. Graduate studies are not simple factual knowledge. How did the PhD experience impact you personally and professionally? Next to these technical skills, my PhD journey instilled in me the crucial value of perseverance. Countless moments of frustration, where results diverged from expectations, tested my resolve. However, I learned to remain committed, viewing setbacks as opportunities for learning and growth. This unwavering resilience is a valuable asset that I carry not only in research but in all aspects of my life. Can you tell us about some of your favorite moments? My PhD journey was woven with memorable moments that fostered growth and connection. My first-ever improv experience at Second City improv after my first RSNA, with Dr. Samei, Dr. Ria, and colleagues where performers hilariously targeted me, cemented the memory! Engaging with brilliant minds at conferences like RSNA, AAPM, and SPIE broadened my horizons, while reconnecting with former colleagues thriving in industry and academia sparked joy. Besides, the sunrise run with Dr. Samei at AAPM at DC, CVIT’s 5K run at “angle among us” and being “unofficial” member of the CVIT soccer team were unforgettable memories I have. And during all of this 4 yours of research, as an international student navigating a new country, Dr. Samei’s unwavering support was a constant anchor during challenging moments. The graduated neuron doll he gifted me right after my defense will always be a cherished reminder of these times!
The Center for Virtual Imaging Trials will be, for the first time, at the RSNA annual meeting in Chicago. From November 26-29, CVIT will be at McCormick Place, booth 3352, First Time exhibitor Pavilion, South Hall Level 3. At the conference, the Center will showcase its activities and resources. It will also be possible to meet CVIT leaders and team members, and to schedule one-to-one meetings. CVIT also presents the Virtual Imaging Trials in Medicine (VITM24) summit that will be hosted at Duke University, Durham (NC), April 22-24, 2024. The summit will gather leading experts, researchers, and practitioners in medical imaging and therapy using in silico virtual trials and digital twins in medicine. VITM24 will include presentations by leading scientists, practitioners, and developers of related scholarship; workshops; proffered presentations, and industry expo. As usual, together with RAI Labs and CIPG members, CVIT researchers will present several scientific abstracts covering topics including quality assurance in radiology, CT radiation dose and image quality monitoring, clinical and radiation risk assessment in CT, and quality improvement in MRI. CVIT Director Dr. Ehsan Samei will also lead a session about virtual clinical trials and its tole in radiology. Following is the list of presentation: Scott Robertson: The power of PI-QUAL metrics and multidisciplinary collaboration: building a robust and comfortable non-endorectal prostate MRI protocol. November 27th, 8:00am, room S405 Njood Alsaihati: Radiology professional preferences for CT radiation dose and image quality monitoring. November 27th, 12:45pm, Learning Center. Nicole Lafata: Practical implementation of radiography quality assurance. November 27th, 1:30pm, room N229. Scott Robertson: More with less: a quality improvement initiative to evolve multiparametric MR prostate imaging beyond the endorectal coil. November 28th, 12:45pm, Learning Center. Ehsan Samei: Virtual Clinical trials. November 28th, 4:30pm, room S404. Francesco Ria: Clinical and radiation risk across one million patients in computed tomography: influence of age, size, and race. November 30th, 1:15pm, Learning Center.
The Submission Deadline is approaching – Act Now! Don’t miss the chance to be part of the first VITM conference! This is your opportunity to position your groundbreaking research in virtual imaging at the forefront of the medical imaging industry. Abstract Submission Deadline: October 20th Visit this page to submit your abstract. International Summit in Virtual Imaging Trials 2024 Hosted at Duke University, the VITM24 conference is a gateway to the latest advancements in virtual trials for medical imaging and beyond. We bring together leading experts, researchers, and practitioners in the field for an event like no other. Special sessions We are excited to announce the VITM24 special sessions designed to enhance the conference experience. The intent is to foster meaningful discussions and practical applications in the medical field through: Deployment Sessions: Dive into the practical aspects of implementing cutting-edge medical technologies and research findings. Learn strategies from industry, share challenges, and explore best practices for successful deployments. Discussion Sessions: Engage in interactive discussions with fellow attendees and experts. These sessions cover a wide range of topics related to our conference theme, offering a platform for sharing insights, asking questions, and igniting conversations that drive innovation. During our Discussion Sessions, there will be the opportunity to delve into the following thought-provoking topics: Hybrid, real and virtual trials: how do they connect? Why have virtual trials not yet replaced clinical trials? Why are we not yet doing VITs in MRI, US, or other imaging types? Gaps in the modeling space? Qualification of tools and justification of use? Bridging towards simulation of metabolism, treatment, disease progression: basic science processes? Can VITs improve diversity in the “patients” tested? How? How do we generate digital representations of patients for which we have less/no medical information? Visit our program to know more about the event. A Fusion of Research and Clinical Expertise VITM24 isn’t just a conference, it’s a dynamic convergence of research and clinical excellence. Explore the latest research findings, methodologies, and innovations in the realm of medical imaging and therapy. But also discover practical applications, experiences, and insights related to healthcare, medical practices, and clinical trials during clinical sessions.
A new important scientific event is approaching soon involving several Center for Virtual Imaging Trials members. Together with the Clinical Imaging Physics Group (CIPG) and RAILabs colleagues, CVIT researchers are heading to Houston, TX for the 2023 American Association of Physicist in Medicine (AAPM) annual meeting. Between July 23-27, thousands of medical physicists and healthcare professionals from around the World will gather to discuss the latest advancement in the field with CVIT, CIPG, and RAILabs presenting a total of 15 scientific studies. The theme of the conference “The Art of Science, the Science of Care” has been proposed by 2023 AAPM President and CVIT Director Dr. Ehsan Samei. Throughout the annual meeting, he will lead several Presidential events together with prominent artists and thought leaders. In his letter to the medical physics community, Dr. Samei highlighted how the theme of the conference: “reminds us of the privilege of being a medical physicist, a vocation rooted in ingenuity, imagination, and creativity (the art, if you will), which is then enacted within a field that is one of the highest embodiments of human intelligence (physics), and all that, then translated into a practice with direct benefit to human health.” The following is a list of the CVIT presentations. Stay tuned to our social media for more details and to follow CVIT members during their presentations. Nicole Lafata: Clinical Evaluation of AEC Target Exposure Accuracy in Radiography: A Generalized Methodology across Diverse Systems. July 23rd, 2:14pm, Room 371 (GRBCC). Francesco Ria: Comparative Risk Assessment of Clinical and Radiation Risk across a Cohort of Patient and Individualized Risk Optimization. July 23rd, 3:00pm, Exhibit Hall Forum 8 (Level 1, GRBCC). Francesco Ria: Accuracy of Noise Magnitude Measurements from Patient CT Images: A Virtual Imaging Study. July 23rd, 3:30pm, Exhibit Hall Forum 8 (Level 1, GRBCC). Madhura Khandekar: Training an Algorithm to Segment Motion Artifacts in CT Using Virtual Imaging Data. July 23rd, 3:30pm, Exhibit Hall Forum 8 (Level 1, GRBCC). Ehsan Samei: President’s Symposium: The Art of Science, the Science of Care. July 24th, 10:15am, Assembly BC (Level 3, GRBCC). Steven T. Bache: What Is the Optimum kV for Photon Counting CT Imaging of Adults and Children? July 24th, 1:15pm, Exhibit Hall Forum 8 (Level 1, GRBCC). Ehsan Samei: President Symposium Fireside Chat. July 24th, 3:15pm, Room 340 (GRBCC). Amar Kavuri: A Dockerized CT Simulator with a User-Friendly Graphical User Interface: Development and Initial Demonstration. July 25th, 9:30am, Exhibit Hall Forum 8 (Level 1, GRBCC). Nicholas D. Felice: Photon-Counting CT Compared to Energy-Integrating CT for Detection of Liver Lesions. July 25th, 2:15pm, Room 361 (GRBCC). Mridul Bhattarai: Edge-on Irradiated Silicon-Based Photon-Counting CT Vs. Energy-Integrating CT for Bronchitis Quantification: A Virtual Imaging Trial Study. July 25th, 2:25pm, Room 361 (Level 3, GRBCC). Raj Kumar Panta: Liver Fat Quantification with an Edge-on-Irradiated Silicon Photon-Counting CT in a Virtual Imaging Trial. July 25th, 2:35pm, Room 361 (Level 3, GRBCC). Ehsan Samei: Medical Physics Meets Visual Arts. July 25th, 4:00pm, Innovation Room 1 (Level 1, GRBCC). Hananiel Setiawan: Development and Testing of a Clinical Tool to Predict and Optimize Liver Contrast-Enhanced CT Imaging. July 26th, 9:30am, Exhibit Hall – Forum 8 (Level 1, GRBCC). Ehsan Samei: Presidential Forum: Medical Physics NOW. July 26th, 4:30pm, Room 372 (GRBCC). Isabel S. Montero: Sensitivity of Trial Results on the Simulation Parameters in Virtual CT Imaging Trials. July 27th, 7:50am, Room 361 (GRBCC).
CVIT is a partner of the new Research Triangle Center of Excellence in Regulatory Science and Innovation
Great news for Duke University and CVIT as they become partners of the new Research Triangle Center of Excellence in Regulatory Science and Innovation (Triangle CERSI), officially approved by the U.S. Food and Drug Administration (FDA). Duke University and the University of North Carolina at Chapel Hill joined together to create the new Center that will receive up to $50 million from FDA in the next five years. The Center also includes collaborations with North Carolina State University and North Carolina Central University. Triangle CERSI becomes the fifth Center in the U.S. operating under the FDA CERSI Program led by the Office of Regulatory Science and Innovation. The goal of the program is to promote innovation in regulatory science to support FDA’s regulatory science needs. The Center for Virtual Imaging Trials will contribute to the new Triangle CERSI activities offering expertise and technologies for conducting human trials of imaging methods by applying virtual imaging techniques. Such tools developed by CVIT scientists, enable computational simulation comparisons with known ground truth and represent a unique resource in the current regulatory framework. In this scenario, “the Triangle CERSI is a significant opportunity for our scholarly communities to curate and direct our intelligence towards addressing an important societal need for proficient and efficient regulatory approval and oversight”, said Dr. Ehsan Samei, one of the three principal investigators from Duke University and CVIT Director. The Center for Virtual Imaging Trials is proud to be part of the new Triangle CERSI.
Hosted at Duke University, this summit will gather leading experts, researchers, and practitioners in medical imaging and therapy for in silico virtual trials and digital twins in medicine.
Two Center for Virtual Imaging Trials researchers will present their scientific works at the 2023 International Conference of the American Thoracic Society (ATS) that will be held in Washington, DC 19-24 May. Amar Kavuri and Saman Sotoudeh-Paima will present their studies concerning the application of virtual imaging techniques for emphysema quantification and score in chronic obstructive pulmonary disease (COPD). Every year, the ATS International Conference registers the participation of nearly 14,000 pulmonary, critical care, and sleep professionals from around the word. The application to pulmonary diseases is one of the many potential translational opportunities that virtual imaging techniques offer in the medical field. In particular, Saman Sotoudeh-Paima study exploits virtual imaging datasets to inform artificial intelligence algorithms assessing emphysema score for COPD quantifications in CT. Amar Kavuri study won the ATS 2023 abstract scholarship and applies virtual imaging tools to assess the effects of intra-patient end-inspiration variability in emphysema quantification. Following are the details of the presentations. A. Kavuri, M. Nejad, S. Sotoudeh-Paima, H. P. McAdams, D. A. Lynch, P. W. Segars, E.Samei, E. Abadi; Effects of Intra-patient End-inspiration Variability in Emphysema Quantification: A Virtual Imaging Study. American Thoracic Society, 2023. https://www.atsjournals.org/doi/pdf/10.1164/ajrccm-conference.2023.207.1_MeetingAbstracts.A4022 The purpose of this study was to quantify the effects of lung respiration levels on emphysema quantification using a validated virtual imaging platform and to demonstrate the effectiveness of the current adjustment methods. Our results show 15% increased by 46 HU (95% CI:[36.4,56]) per 1L respiration volume deviation and LAA-950 reduced by 1.27% /1L (95% CI:[0.067,2.5] ) deviation. The physiologic based adjustment model reduced this variability more compared to the statistical one due to its basis on the individual patient and not population averages. This analysis helps in improving the lung volume correction methods and in reducing this variability for accurate estimation of the quantitative metrics. Sotoudeh-Paima, S., Nejad, M. G., Segars, W. P., O’Sullivan-Murphy, B., Macintyre, N. R., Lynch, D. A., Samei E., Abadi, E. Emphysema Score in CT for COPD Quantifications Using Artificial Intelligence Informed by Virtual Imaging Datasets. American Thoracic Society, 2023. https://doi.org/10.1164/ajrccm-conference.2023.207.1_MeetingAbstracts.A4021 Computed tomography (CT) is an in vivo diagnostic method that assesses the severity and extent of emphysema in the lungs. LAA-950 is a conventional imaging biomarker to quantify emphysema. While this biomarker has shown promising values in assessing disease severity, it is highly prone to variability in imaging protocols and scanner makes and models. The purpose of this work was to investigate a deep learning (DL)-based quantification approach that can characterize emphysema accurately while being robust to scanner variability. This approach would enable a more reliable comparison of emphysema quantifications in multi-center and longitudinal studies.
CVIT member Saman Sotoudeh-Paima received finalist for the Robert F. Wagner award at SPIE Medical Imaging conference
Last week, CVIT member Saman Sotoudeh-Paima received the Finalist for the Robert F. Wagner all-conference best student paper award at SPIE Medical Imaging conference held in San Diego, California. The SPIE medical imaging chapter is focused on the latest advances in image processing, physics, computer-aided diagnosis, perception, image guided procedures, biomedical applications, ultrasound, informatics, radiology and digital computational pathology. Saman, togethers with co-authors Dr. Ehsan Abadi and Dr. Ehsan Samei, presented a study about a new CT imaging harmonizer providing robust pulmonary emphysema quantifications and enabling objective disease characterizations in large-scale, multi-center, and longitudinal studies. In particular, pulmonary emphysema is a form of Chronic Obstructive Pulmonary Disease (COPD) and a chronic lung condition that results in a breakdown of alveoli walls. Quantitative Computed Tomography is increasingly used to assess the presence or progression of emphysema. However, CT quantifications are affected by the acquisition protocols and scanner makes and models. This variability is a major concern for cross-sectional and longitudinal disease characterizations with largescale, multi-institutional datasets. Therefore, CT images need to be harmonized to reflect the patient condition and not the attributes of the imaging systems. The proposed new framework, named CT-HARMONICA, was developed using a virtual imaging trial (VIT) platform at the Center for Virtual Imaging Trials. CT-HARMONICA transforms CT images to a reference quality index (iso resolution and noise conditions) enabling robust emphysema quantifications across varied CT conditions. The developed harmonizer was applied to clinical data from the COPDGene dataset to demonstrate its clinical utility. The established imaging biomarkers of LAA-950 and Perc15 were selected for emphysema quantifications. Results demonstrated that the harmonizer improved the quantification performance by reducing the bias in LAA-950 from 7.03 (CI: [6.38, 7.68]) to 0.14 (CI: [0.08, 0.20]) after matching for kernel and from 2.48 (CI: [2.21, 2.76]) to −0.34 (CI: [−0.48, −0.20]) after matching for noise settings on the COPDGene dataset. Saman is a second year Ph.D. student at the Department of Electrical & Computer Engineering, Pratt School of Engineering, Duke University. Currently, he is a Research Assistant at Carl E. Ravin Advanced Imaging Laboratories (RAI Labs) and the Center of Virtual Imaging Trials (CVIT). His research is primarily focused on improving the accuracy and precision of CT quantification for COPD patients, mainly using image processing- and deep learning-based techniques.