Newsletter.

Welcome to our newsletter!

Sign up to never miss an issue. Please also check out prior issues.

*|MC:SUBJECT|*
Welcome to the July 2024 issue of the
Medical Imaging and Data Resource Center (MIDRC) newsletter!

MIDRC Helper AI and Pixel DeID:
Enhancing Medical Imaging Analysis with AI

The MIDRC team is committed to advancing the field of medical imaging through innovative technology and thorough data preparation. Recently, we've developed two AI models designed to enhance the accuracy and privacy of imaging data: MIDRC Helper AI and Pixel DeID.

The MIDRC Helper AI model improves the accuracy of imaging metadata by annotating data with correct anatomical information, orientation, and contrast. For example, Helper AI can effectively address issues like mislabeled studies, which can significantly improve research outcomes. By deploying this model, we can enhance both the reliability and usability of medical images.

Pixel DeID addresses the challenge of de-identifying pixel-level data in medical images. This model automatically detects sensitive information burned into the images. The detected pixel regions can then be automatically redacted. For example, Pixel DeID + Redaction can automatically detect protected health information (PHI) like patient names burned into the image and redact them. This automated tool not only saves time, but also minimizes the risk of human error. It ensures that medical images are shared for research purposes without jeopardizing patient privacy.

These tools align with MIDRC’s mission to provide high-quality, de-identified medical imaging data for research. As we continue to refine these models, our focus will expand to include a wider variety of imaging modalities and conditions. For in-depth demonstrations and to see these models in action, please check out the
video on our YouTube channel.

Pixel DeID and Redaction: These images showcase the Pixel DeID model, which addresses the challenge of de-identifying pixel-level data in medical images. The first image illustrates the detection of protected health information (PHI) such as patient names, while the second image shows the result after automated redaction. This tool ensures that medical images are shared for research purposes without jeopardizing patient privacy. *Please note, no actual PHI is included in the image*
MIDRC Helper AI in Action: The first image shows the Helper AI model's output on a CT scan of the abdomen. This AI model improves imaging metadata accuracy by adding correct anatomical information, orientation, and contrast, fixing mislabeled studies and enhancing research outcomes.
For in-depth demonstrations and to see these models in action, visit our YouTube channel, MIDRC Media, to watch these informative videos

MIDRC Diversity Calculator:
Assessing the Representativeness of Biomedical Data

The MIDRC Diversity Calculator enables researchers to assess the representativeness of biomedical data. This open-source tool uses the Jensen-Shannon distance (JSD) measure to evaluate demographic representativeness, ensuring that biomedical data reflects diverse populations.

Key Features of the Diversity Calculator:

  1. Jensen-Shannon Distance (JSD) Calculation: Assesses how closely data aligns with target population demographics.
  2. Comparative Analysis: Compares different datasets to highlight diversity gaps, a crucial feature for ensuring equity in biomedical research.
  3. Historical Data Monitoring: Tracks changes in data representativeness over time, enabling researchers to monitor changes and track progress toward more inclusive datasets.
  4. Biomedical Focus: Specifically tailored for analyzing biomedical data (both imaging and non-imaging), ensuring that comparisons are relevant and meaningful within the context of health research. 
  5. Generalizability: Adaptable to various datasets and demographics, with easy data input via Excel.
Practical Applications and Future Developments
The Diversity Calculator, initially used for COVID-19 research, now extends to cancer and neurological conditions, supporting demographic comparisons. Its ability to analyze data in real time via APIs will be demonstrated in the coming weeks. By comparing populations and their attributes, the Diversity Calculator can be instrumental in identifying equity gaps in research on diseases such as cancer. Such findings will inform recruitment priorities for future clinical studies, ensuring a more inclusive and representative approach to biomedical research.

How the Diversity Calculator Aligns with MIDRC's Objectives
The Diversity Calculator is a testament to MIDRC's commitment to advancing research through science-based, inclusive approaches. By providing a tool that describes the representativeness of biomedical data, the Diversity Calculator supports MIDRC’s overarching goals:
  1. Expanding AI-Ready Data Commons: Ensuring representative data for a well-curated, diverse repository following FAIR principles (Findable, Accessible, Interoperable, and Re-usable).
  2. Fostering Machine Intelligence Research: Facilitating the development of trustworthy algorithms for disease-related AI applications by providing diverse datasets.
  3. Supporting Data Management and Sharing: Enhancing data quality and applicability, in line with the NIH Data Management and Sharing Policy.

Conclusion
The MIDRC Diversity Calculator is crucial for promoting inclusive and representative biomedical AI/ML research. By ensuring data diversity, MIDRC advances research and health equity.

For more information and to access the Diversity Calculator, visit
MIDRC's website. The software is also available on the MIDRC GitHub pageand you can find a link to our algorithms page here

1st place - $15,000

2nd place - $8,000

3rd place - $7,000

4th-7th place (each) - $5,000
 

Registration opens July 19, 2024

Calibration phase opens August 10, 2024

Validation phase opens September 10, 2024

Test phase opens October 1, 2024

Challenge concludes October 22, 2024

Winners Announced early November, 2024!

Registration Open!
Please review the MIDRC Grand Challenge Conflict Policy for participation eligibility

MIDRC at the AAPM Annual Meeting:
Empowering Researchers with Open-Source Tools

MIDRC remains committed to making its tools and data accessible to all researchers. At AAPM’s Annual Meeting in Los Angeles from July 21-25, attendees visited the MIDRC Booth (#606) to learn about our open-source resources:

  • MIDRC MetricTree: An interactive tool recommending performance metrics for computational research.
  • MIDRC Bias Awareness Portal: A resource to understand and mitigate bias in AI/ML.
  • MIDRC Diversity Calculator: A tool for measuring the representativeness of biomedical datasets.
  • Cohort Selection in the MIDRC Data Commons: Algorithms and methods for multi-modal investigations.

We were excited to share our advancements in medical imaging research and hope to see you next year! For more information, visit MIDRC at AAPM Annual 2024

Pictured: Dr. Curtis Langlotz, MIDRC PI and RSNA President, speaking at the AAPM President’s Symposium during the 2024 Annual Meeting in Los Angeles, and multiple photos of MIDRC researchers showcasing cutting-edge resources at the MIDRC booth.

Featured Article:
CheXpert Plus - Enhancing Medical Imaging Research
 
A team led by Dr. Curt Langlotz of Stanford University announces the release of CheXpert Plus, an extensive dataset designed to advance medical imaging research. CheXpert Plus includes a comprehensive collection of de-identified chest radiograph reports, demographics, concept extractions, and DICOM images. This dataset expands on the original CheXpert images and incorporates additional data formats and annotations linked back using privacy-preserving record linkage.
 
Developed through a collaborative effort involving MIDRC-funded researchers, CheXpert Plus serves as a pivotal resource for advancing AI technologies in medical imaging. It provides substantial data for training and validating AI models, enhancing research capabilities across numerous applications. A detailed manuscript outlining the dataset has been published on arXiv, with a submission to a high-impact journal underway.
 
Key Resources:
For more information, please visit Stanford AIMI's Shared Datasets.

The Medical Imaging and Data Resource Center (MIDRC) is a collaborative initiative leveraging AI to revolutionize medical imaging, enhance diagnostic accuracy, and improve patient care.

Our virtual seminars offer the medical community a chance to hear from our team. Free and open to everyone! We will resume in September, as no seminars are held over the summer months.

 
Connect with us!
Follow MIDRC Data Commons on LinkedIn to stay informed on the latest updates & news about our research activities and advancements

Stay tuned for the latest updates & innovative developments in medical imaging research! Follow MIDRC Media on YouTube @MIDRC_
As MIDRC expands into oncology data collection, you can help by providing data from your institution or hospital. For questions or to discuss contributions, please contact us here

If you have technical questions regarding data and/or annotations available in the MIDRC data portal, please contact midrc-support@datacommons.io

Our growing data portal is live and can be accessed data.midrc.org

Our Quick Reference Guide for Downloading Image Data from the MIDRC Gen3 Data Commons can be found
here

If you are interested in becoming a MIDRC partner please contact erin.mueller@bsd.uchicago.edu

Find us on our YouTube Channel, Twitter, LinkedIn, and GitHub to stay informed about all things MIDRC and sign up for our newsletter!