Improving fairness in machine learning for radiology diagnostics

SCH: Quantifying and mitigating demographic biases of machine learning in real world radiology

['FUNDING_R01'] · JOHNS HOPKINS UNIVERSITY · NIH-10875636

This study is looking at how computer programs used in medical imaging might unfairly affect certain groups of people, and it aims to make sure that cancer screenings, like those for breast and lung cancer, are accurate and fair for everyone, no matter their background.

Quick facts

Phase['FUNDING_R01']
Study typeNih_funding
SexAll
SponsorJOHNS HOPKINS UNIVERSITY (nih funded)
Locations1 site (BALTIMORE, UNITED STATES)
Trial IDNIH-10875636 on ClinicalTrials.gov

What this research studies

This research investigates how machine learning algorithms used in radiology can unintentionally produce biased results against under-represented demographic groups, which may worsen existing health disparities. The project aims to develop methods to quantify and correct these biases, ensuring that diagnostic tools are fair and equitable for all patients. By analyzing how these algorithms perform across different demographic groups, the research seeks to enhance the accuracy and accessibility of cancer screening, particularly for breast and lung cancer. The approach includes creating algorithms that can adjust for biases even when sensitive demographic information is not directly available.

Who could benefit from this research

Good fit: Ideal candidates for this research include individuals undergoing breast or lung cancer screening, particularly those from under-represented demographic backgrounds.

Not a fit: Patients who are not involved in cancer screening programs or those from demographic groups that are already well-represented in existing data may not benefit from this research.

Why it matters

Potential benefit: If successful, this research could lead to more accurate and equitable cancer screening tools that benefit all demographic groups.

How similar studies have performed: Other research has shown promise in addressing algorithmic bias in healthcare, indicating that this approach could lead to significant advancements in equitable medical diagnostics.

Where this research is happening

BALTIMORE, UNITED STATES

Researchers

About this research

  1. This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
  2. Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
  3. For full project details, budget, and progress reports, visit the official NIH RePORTER page below.

View on NIH RePORTER →

Conditions: Breast Cancer, Breast Cancer Detection, Breast cancer screening

Last reviewed 2026-05-15 by the Find a Trial editorial team. Information on this page is for educational purposes and is not medical advice. Always consult qualified healthcare professionals about clinical trial participation.