Vega President/CEO Featured in Radiology Today

Vega Imaging Informatics President and CEO, Luke Bideaux, was featured in the March/April issue of Radiology Today, sharing lessons learned from, along with the significance of, the company’s recent completion of the world’s largest Digital Breast Tomosynthesis (DBT) dataset with paired histology outcomes.

This dataset included over one million DBT studies across three DBT manufacturers (equating to over 200 million images when broken down by frame), which were paired with biopsy outcome data for over 22,000 patients, containing over 7,000 cancer cases.

Why was this project so significant?

1) The Gold Standard: Pairing DBT images with ground-truth histology is essential for training robust, clinically relevant AI models for cancer detection and prediction. This pairing allows for the creation of gold-standard labels that eliminate ambiguity in 3D lesion classification.

2) The Size: As Luke discusses in the article, this dataset was significant due to its large size, nearly 1 petabyte of data, as it was intended to represent a large distribution of demographics, breast densities, BI-RADS scores, and cancer types, among other criteria, all based on real-world occurrences.

3) The Future of AI Model Development in Breast Imaging: This project also holds great significance for the future of AI in breast imaging. This dataset was used for development of an AI model for breast imaging, meaning that it and future solutions will work more effectively because broader patient demographics were represented in the data that was used for training and testing this model.

4) The Need for Real World Data: Lastly, this project holds great significance for healthcare organizations, including hospitals and out-patient imaging centers, who make these types of projects and the future of AI model development possible. As Luke states in the article, “I think the message is pretty clear from the healthcare providers that they need AI that actually works in the real world, and the only way that’s going to happen is if we get more healthcare providers contributing data for the advancement of medical imaging AI.”

Read the full feature here: March/April 2026 – Radiology Today

Breast Imaging AI Developers interested in obtaining DBT datasets for development of their AI models and/or healthcare organizations interested in monetizing their data for the future of AI model development can schedule a one-on-one meeting with Vega here: https://calendly.com/alkappel