by
Gus Iversen, Editor in Chief | October 23, 2025
HOPPR has released a new Vision Transformer-based foundation model aimed at accelerating artificial intelligence development in mammography.
The company said the HOPPR EB 2D Mammography Foundation Model is designed to support fine-tuning for a range of imaging tasks, including cancer detection, breast density classification, and device identification.
The model, trained using self-supervised learning and expert distillation, is delivered with LoRA adapters and a lightweight MLP head. According to the company, it returns structured predictions in JSON format and is intended to integrate easily into AI development workflows using labeled DICOM data.
Ad Statistics
Times Displayed: 5266
Times Visited: 18 Brand-New FDA-cleared Advanced Ultrasound Medical Device available for sale or lease to Wound Care Centers or any other Medical Facilities.The Arobella 1000D is designed for non-contact or debridement ultrasound wound healing therapy, or any other wounds
The release adds to HOPPR’s growing model library, which includes its chest radiography foundation model, and is available through a secure API. The platform supports HIPAA-compliant inference, usage-based billing, and traceable development environments built under a quality management system aligned with ISO 13485, SOC 2, and HITRUST standards.
“Foundation models are changing the pace of innovation in imaging AI, but only if they’re accessible, adaptable, and built with real-world deployment in mind,” said Dr. Khan Siddiqui, CEO and cofounder of HOPPR. “With this release, we’re giving developers the infrastructure to move quickly with transparency, traceability, and control from day one.”
Internal testing reported an area under the curve (AUC) of 0.92 for cancer detection, 0.94 for breast density classification, and 0.99 for pacemaker detection. While these figures reflect internal validation, the company did not specify the data sets used or whether external benchmarking has been conducted.
Sham Sokka, chief operating and technology officer at DeepHealth, said the foundation model “should give us a flexible infrastructure to adapt it to our workflow needs.”
The announcement coincides with Breast Cancer Awareness Month and comes ahead of a webinar hosted by HOPPR and DeepHealth on October 23, focused on the evolving role of foundation models in medical imaging.
