Overview

Schema Version

https://atlas.rsna.org/schemas/2025-11/dataset.json

Name

iBRISK multicenter BI-RADS 4 mammography and clinical dataset

Link

https://pubmed.ncbi.nlm.nih.gov/38074778/

Indexing

Keywords: Mammography, Breast, Oncology, Biopsy/Needle Aspiration, Radiomics, Precision Mammography, AI-augmented Biopsy Decision Support Tool, Breast Cancer Risk Calculator, BI-RADS 4, Risk Stratification, Probability of Malignancy, PPV3

Content: BR, OI

RadLex: RID35976, RID36039, RID12790, RID12933, RID49911, RID36030, RID45700, RID10357

Author(s)

Chika F. Ezeana

Tiancheng He

Tejal A. Patel

Virginia Kaklamani

Maryam Elmi

Erika Brigmon

Pamela M. Otto

Kenneth A. Kist

Heather Speck

Lin Wang

Joe Ensor

Ya-Chen T. Shih

Bumyang Kim

I-Wen Pan

Adam L. Cohen

Kristen Kelley

David Spak

Wei T. Yang

Jenny C. Chang

Stephen T. C. Wong

Organization(s)

Houston Methodist Hospital (Houston Methodist Neal Cancer Center)

MD Anderson Cancer Center

University of Texas Health Science Center San Antonio (UTHSCSA)

License

Text: CC BY 4.0

URL: https://creativecommons.org/licenses/by/4.0/

Contact

Corresponding author: Stephen T. C. Wong, Department of Radiology, Houston Methodist Hospital, Weill Cornell Medicine, 6670 Bertner Ave, Houston, TX 77030; email as listed in article.

Funding

Supported by the Ting Tsung & Wei Fong Chao Family Foundation, the John S. Dunn Research Foundation, the Breast Cancer Research Foundation, and NIH/NCI grant no. 1R01CA251710.

Ethical review

IRB-approved at participating institutions; HIPAA compliant; waivers of informed consent were granted.

Comments

Retrospective, multi-institutional dataset of women with BI-RADS category 4 lesions used to assess the iBRISK decision support model. Includes patient clinical factors and mammography report descriptors with biopsy outcomes as ground truth.

Date

Published: 2023-08-09

References

[1] Ezeana CF, He T, Patel TA, et al.. "A Deep Learning Decision Support Tool to Improve Risk Stratification and Reduce Unnecessary Biopsies in BI-RADS 4 Mammograms". Radiology: Artificial Intelligence. 2023-01-01. doi:10.1148/ryai.220259. PMID: 38074778. PMCID: PMC10698614.

[2] He T, Puppala M, Ezeana CF, et al.. "A deep learning-based decision support tool for precision risk assessment of breast cancer". JCO Clinical Cancer Informatics. 2019-01-01. PMID: 31141423. PMCID: PMC10445790. Available from: https://pubmed.ncbi.nlm.nih.gov/31141423/

Dataset

Motivation

Assess iBRISK performance to improve risk stratification of BI-RADS 4 lesions and reduce unnecessary biopsies and associated costs.

Sampling

Consecutive case curation within stated time windows at each institution; inclusion required BI-RADS 4 at diagnostic mammography and biopsy within 3 months, with specified minimum data fields.

Partitioning scheme

Model originally developed on 9700 patient records and validated on 1078 patients from the primary institution; multicenter independent retrospective test set of 4209 patients from three institutions.

Missing information

UTHSCSA had 1.34% missing data across features; HMH and MDACC validation data were complete per report.

Relationships between instances

Each patient record links clinical factors and mammography report descriptors to biopsy-proven pathologic outcome (benign vs malignant).

External data

Data curated consecutively from three Texas institutions: HMH (systemwide data warehouse), MDACC (EMR), UTHSCSA (EMR).

Confidentiality

Retrospective, IRB-approved, HIPAA-compliant; waivers of informed consent.

Sensitive data

Clinical and report data derived from EMR/data warehouse; handled under HIPAA; not publicly released in the article.