Columbia UniversityColumbia University
Technologies
TechnologiesProductsCollectionsChat with AI
Sign in
Columbia University

Automated risk assessment for breast cancer

OverviewRequest InfoMore Like ThisProducts

Products

LIBRA

Tags

Case–control studyMammographyRadiologyApplied mathematicsRadiodensityBreast cancer screeningScalabilityDensity estimationBreast cancer

Inventors

Brad KellerDespina Kontos

Departments

Radiology

Divisions

Columbia University Medical Center (CUMC)

Manager

Joan Martinez

File Number

CU26091

Release Date

Dec 1, 2025

This technology is a fully automated software package that estimates breast density from digital mammographs for breast cancer risk prediction models.

Unmet Need: Automated assessment of digital mammograph images

Mammographs are used to assess risk for the development of breast cancer by capturing the density of the breast tissue. Analysis of mammograms is performed qualitatively by radiologists or semi-automated methods, which introduces a degree of variability that lowers the predictive value. An automated, reproducible method of analysis of digital mammographs is necessary to maximize the predictive benefit of this risk factor.

The Technology: Automated mammograph analysis for breast cancer screening workflows

The technology, termed the Laboratory for Individualized Breast Radiodensity Assessment (LIBRA), is a fully automated software package that analyzes digital mammography images to generate density measurements. Using recent advancements in applied mathematics, the software segments dense and non-dense breast tissue in mammographs to calculate percent density and generate annotated images. LIBRA offers both interactive and batch modes, demonstrating scalability, robustness, and suitability for both small clinical use and large cohort research.

This technology has been validated on over 70,000 mammography exams.

Applications:

  • Clinical research quality assurance and standardization
  • Data-driven breast cancer risk evaluation and modeling
  • Risk stratification to guide individualized screening strategies

Advantages:

  • Reproducible, standardized density estimation supports multi-site clinical research
  • Compatible across leading mammography platforms
  • User-friendly graphical interface for single mammograph analysis

Lead Inventor:

Despina Kontos, Ph.D.

Related Publications:

  • Habel LA, Alexeeff SE, Achacoso N, Arasu VA, Gastounioti A, Gerstley L, Klein RJ, Liang RY, Lipson JA, Mankowski W, Margolies LR. “Examination of fully automated mammographic density measures using LIBRA and breast cancer risk in a cohort of 21,000 non-Hispanic white women.” Breast Cancer Research. 2023 Aug 6;25(1):92.

  • Gastounioti A, Cohen EA, Pantalone L, Ehsan S, Vasudevan S, Kurudi A, Conant EF, Chen J, Kontos D, McCarthy AM. “Changes in mammographic density and risk of breast cancer among a diverse cohort of women undergoing mammography screening”. Breast Cancer Research and Treatment. 2023 Apr;198(3):535-44.

  • Keller BM, Chen J, Daye D, Conant EF, Kontos D. “Preliminary evaluation of the publicly available Laboratory for Breast Radiodensity Assessment (LIBRA) software tool: comparison of fully automated area and volumetric density measures in a case–control study with digital mammography.” Breast cancer research. 2015 Aug 25;17(1):117.

Tech Ventures Reference:

  • IR CU26091

  • Licensing Contact: Joan Martinez

Discover More with AI