MIDRC newsletter.

Newsletter.

Stay in touch through the MIDRC newsletter! Sign up to receive the newsletter.

Issue 7, January 2023:

*|MC:SUBJECT|*

January 2023

In just two years, we have released over 100,000 imaging studies to the public to advance the development of ML/AI for COVID-19. 

Visit the public data portal, data.midrc.org

In addition to the public data portal, MIDRC has a sequestered dataset to facilitate regulatory clearance and accelerate clinical usage. Methods of selecting clinical cases for independent testing are used to appropriately evaluate a developed algorithm.  
The RSNA 2022 Annual Meeting was held in-person in Chicago, IL from
Nov. 27 to Dec. 1. This convention highlighted the latest in medical imaging research and education and emphasized the importance of empathy, diversity and equity in health care. The meeting’s theme was “Empowering Patients and Partners in Care.” MIDRC had a booth that was well-attended by investigators and vendors! Check out pictures from this year's meeting. If you weren't able to attend this year,
we hope to see you next fall. 
AAPM 2023 Abstract submission

The AAPM's upcoming 65th Annual Meeting & Exhibition is the world's largest program of scientific, educational, and professional presentations and exhibits in the medical physics industry, drawing thousands of attendees! The electronic abstract submission portal is now open and accepting 300 word proffered abstracts and supporting data. Abstract submission, important dates and all meeting information can be found here.

Submissions open: Jan. 19, 2023; Noon CT

Submissions close: February 28, 2023;  7PM CT

RSNA 2023 Abstract submission

Feeling inspired by the scientific and educational content from RSNA 2022?
Become a presenter at RSNA 2023! They are looking for thought leaders to provide scientific presentations, education exhibits and quality improvement reports. 
Abstract submission, important dates and all meeting information can be found here.

Submissions open: Feb. 23, 2023; Noon CT

Submissions close: May 3, 2023; Noon CT

In preparation for our next Grand Challenge, we launched a crowdsourcing annotation event, ‘Annotate Away COVID’. The event kicked-off at RSNA 2022 and is continuing as a virtual event. This COVID severity annotation event on chest radiographs utilizes publicly available MIDRC imaging studies and seeks to build an annotated training set and helps provide a reference standard for AI algorithms!

The annotations collected will serve, after curation, as the ground ‘truth’ of the training set for our next Grand Challenge on COVID severity and will be available to researchers using MIDRC data at large. This event, annotating frontal chest radiographs for COVID severity, is going on right now! You can take part in this event on your own laptop/computer and the process is fast and straightforward.
Anyone can participate!

Yes, I want to help!

Seminar Series | Virtual Event
Thank you to everyone who could join us for our conversation with Dr. Curtis Langlotz, Dr. Tessa Cook and Pierre Chambon discussing, 
Natural Language Processing of Clinical Text for COVID-19 Identification

A full video recording of the Seminar will be posted to our free 
MIDRC YouTube channel. Please subscribe to access this and all past MIDRC presentations.
Our next Seminar is scheduled for Tuesday, February 21 at 2pm CT,
and will feature MIDRC Principal Investigator Dr. Paul Kinahan (University of Washington) discussing "Curating the MIDRC collection of DICOM images for use in AI research". 
Seminar Registration
MIDRC Researcher Spotlights 

Berkman Sahiner, MS, PhD
 
Berkman received his BS and MS degrees from the Middle East Technical University in Ankara, Turkey, and his PhD degree from the University of Michigan, Ann Arbor, all in Electrical Engineering and Computer Science. After his PhD, he started as a postdoctoral researcher at the Radiology Department at the University of Michigan. Although his PhD thesis focused on image reconstruction, this postdoctoral position was the first time he was first involved with clinical imaging, and he has been hooked ever since. He became a faculty member at the same department a few years later, and he was an Associate Professor when he moved to the Center for Devices and Radiological Health (CDRH) at the US Food and Drug Administration in 2009.

Berkman works as a Senior Biomedical Research and Biomedical Product Assessment Service Expert at the Office of Science and Engineering Laboratories (OSEL), which is the research arm of CDRH. He is the coordinator for the Artificial Intelligence/Machine Learning (AI/ML) Program, which conducts regulatory science research to ensure patient access to safe and effective medical devices using AI/ML. Berkman has been working in the area of AI/ML applied to imaging since the AI winter in the mid-1990’s, when people were embarrassed to refer to it as AI and called it pattern recognition. In addition to AI/ML, Berkman is interested in image perception, clinical study design, and performance assessment methodologies.

At MIDRC, Berkman is a principal contributor to the technology development project (TDP) 3c research group, which is working on the development of an online tool to guide researchers in selecting appropriate task-based performance metrics, with links to relevant software packages and literature. He is also active in the MIDRC Bias and Diversity Working Group (BDWG), which aims to ensure a strong emphasis on the diversity of MIDRC data and to support the development of fair and generalizable algorithms.

Berkman lives in Rockville, Maryland with his wife, and whenever he visits from graduate school, their son. They are all electrical engineers! Berkman’s wife Ulku works as a customer engagement manager at ACR, and the two of them sometimes find it hard not to speak about MIDRC at dinner table. Berkman is an admirer of the US National Park System and tries to visit one at every new state he goes to. He also likes to solve chess puzzles with his high school buddies online, and wishes that there were more series like the “Queen’s Gambit” to lessen the “nerdy” stigma attached to the game.

Jonathon K. Maffie, MD PhD
 
American College of Radiology® (ACR®) COVID-19 Imaging Research Registry™ (CIRR) site Principal Investigator (PI), Jonathon K. Maffie, MD, PhD, has led Penn State Health Milton S. Hershey Medical Center in contributing 811 images from more than 800 patients to CIRR, which has provided 61,768 images from 27,276 Patients to the Medical Imaging and Data Resource Center (MIDRC) to date.

Maffie’s background in neuroradiology led him to become an active member of the CIRR Steering Committee and a primary COVID-19 researcher. As part of his PhD training, Maffie completed a research project for the National Institute of Neurological Disorders and Stroke on “The Role of DPP6 in Kv4 Localization and Dendritic Excitability.” Having completed a fellowship in neuroradiology at Massachusetts General Hospital, he is pursuing translational research in neuroradiology, including imaging diagnosis of movement disorders.

Maffie is an assistant professor of radiology at the Penn State Health Milton S. Hershey Medical Center. His work has been recognized with a resident/fellow research grant from the Radiological Society of North America (RSNA).

When not working, Maffie enjoys skiing, biking, travel and exploring Hershey, PA with his wife and kids.

Curtis P. Langlotz, MD, PhD

Curtis P. Langlotz, MD, PhD, is chair of the Radiological Society of North America
(RSNA) Board of Directors. A renowned imaging informatics leader and committed advocate for improved clinical communication, Dr. Langlotz is professor of radiology, medicine and biomedical data science, director of the Center for Artificial Intelligence in Medicine and Imaging, and associate chair for information systems in the Department of Radiology at Stanford University. As a medical informatics director for Stanford Health Care, he sets strategy for the computer technology that supports the Stanford Radiology practice.

His biomedical informatics research laboratory aims to reduce diagnostic errors and improve the accuracy and consistency of clinical communication by developing novel AI algorithms that provide real-time assistance to radiologists, clinicians and patients.

A MIDRC principal investigator, Dr. Langlotz has an interest in promoting open data and open science; he sees MIDRC as an unprecedented collaboration of three major North American imaging professional organizations that offers a powerful set of tools and infrastructure to accelerate research not just for COVID-19 but for many other conditions.

A longtime RSNA member, Dr. Langlotz served for many years on RSNA’s Radiology Informatics Committee and has served the Society as an informatics advisor. He led the development of numerous RSNA informatics initiatives, including the RadLex terminology standard, the LOINC-RadLex Playbook of standard exam codes and the RSNA Imaging AI Certificate Program.

He founded and is a past president of the Radiology Alliance for Health Services Research. He received its career achievement award in 2017. Dr. Langlotz served as chair of the Society for Imaging Informatics in Medicine (SIIM) and as a board member of the Association of University Radiologists and the American Medical Informatics Association. He has served on the external advisory board of the National Cancer Institute’s Imaging Data Commons for the past two years.

Dr. Langlotz lives in an “empty nest” in Menlo Park, California, with his wife, who is chair of the Department of Pediatrics at Stanford. He enjoys genealogy, golf, vacationing with his wife and two adult children, and climbing among the redwoods in the foothills around Stanford on his road bike.

More on MIDRC
Read literature published and peer-reviewed by MIDRC investigators regarding
our research.

We are committed to the crucial principles of equity and inclusion. 
Learn more about MIDRC Diversity.

To find out more about the Medical Imaging and Resource Center,
please visit out website!
If you are interested in becoming a MIDRC partner please contact kpizer@bsd.uchicago.edu
Website
YouTube
Twitter
LinkedIn