DeepHealth, a wholly-owned subsidiary of RadNet, Inc. (NASDAQ: RDNT) and a global leader in AI-powered radiology and health informatics, today announces a data and AI development partnership with HOPPR (www.hoppr.ai). This collaboration will commercialize a pioneering Medical-Grade Generalized Foundational Model and foster the development of Fine-Tuned models for breast, prostate, and lung cancer detection, leveraging generative medical imaging-focused AI and robust, diverse data sets.
HOPPR’s Medical-Grade Generalized Foundation Model enhances medical research and hypotheses while simplifying and lowering costs for data collection and training. AI Foundational Models are versatile, pre-trained architectures that serve as a starting point for customizing specific tasks through Fine-Tuned models, for which expertise in a particular domain is critical.
The partnership seeks to create new Fine-Tuned models, powered by HOPPR’s medical-grade foundation model, to bolster DeepHealth’s AI-powered health informatics portfolio by enabling it to create future solutions more quickly and efficiently and to support the evolution of radiology in the coming years. DeepHealth’s cloud-native operating system (OS) is designed to integrate clinical and operational tools, to provide radiology workflow efficiencies and improve patient outcomes.
Sham Sokka, PhD, Chief Operating and Technology Officer, DeepHealth, said, “DeepHealth’s partnership with HOPPR is a significant leap forward in DeepHealth’s mission to empower breakthroughs in care through enabling new diagnostic imaging technologies.”
“The integration of foundational models like those being developed by HOPPR in medical imaging is intended to boost diagnostic accuracy, speed up image analysis, and pave the way for generative AI in non-clinical applications, including workflow automation, ultimately enhancing patient care and outcomes in radiology. At DeepHealth, we are not just a provider of AI technology but are creating a comprehensive portfolio of solutions for medical imaging, seamlessly blending AI-based automation and efficiencies into an operating system for radiology and diagnostic workflows,”
Sham Sokka, PhD, Chief Operating and Technology Officer, DeepHealth
HOPPR’s robust medical-grade infrastructure and tools for accelerating AI and machine learning development, combined with DeepHealth’s deep clinical expertise and successful track record in deploying AI tools at scale and in real-world settings, aim to unlock significant diagnostic, clinical, and operational value from medical imaging data and advance imaging across modalities.
“We are pleased to partner with DeepHealth to transform healthcare informatics,” said Khan Siddiqui, MD – Chief Executive Officer of HOPPR. “Our collaboration on medical grade foundation models and infrastructure supporting them could significantly enhance medical imaging, leveraging AI’s transformative potential to improve clinical care efficiency and quality. HOPPR is collaborating with DeepHealth to build a unified clinical and operational workflow that enables radiologists to efficiently access the information they need through the systems they know.”
DeepHealth’s unique ‘one system’ approach addresses challenges across the entire radiology value chain, from referral management, scheduling, and patient engagement to technologist and radiologist workflows. DeepHealth OS supports radiology departments with a comprehensive solution for medical imaging, including operational solutions and end-to-end services across the care continuum.
DeepHealth technology is used in over 800 clinical sites in select countries, and its AI tools have demonstrated real-world efficacy. For example, its solution for large-scale breast cancer screening programs has the potential to increase the cancer detection rate by up to 18%1. Over fifteen million exams are performed annually using DeepHealth solutions, resulting in more than two million AI-informed diagnoses.
