Segmed and Verily Partner on Real-World Imaging Data

Published February 25, 2026 by The Daily Scout

Segmed, a provider of real-world medical data, has partnered with Verily to expand access to imaging data for research. Segmed's datasets will become available on Verily's AI-native platform. The collaboration aims to support precision health research and the development of new AI models.

Why it matters

- Verily's Pre platform is engineered to harmonize and analyze complex biomedical data, empowering researchers to build and deploy AI models. This AI-native platform processes diverse, multimodal data—both structured and unstructured—to accelerate the development of new therapies. A key feature is its ability to transform static PDF protocols into dynamic digital models, which can significantly reduce the time required for study startups. - The partnership provides researchers with access to Segmed's extensive database of de-identified real-world imaging data, which includes approximately 30 million patients and 150 million imaging exams. An initial dataset available through the collaboration includes 558 digital breast tomosynthesis exams, with 271 of those having biopsy-proven malignant lesions. - To ensure patient privacy, Segmed employs a de-identification process that has received an independent Expert Determination, confirming it meets HIPAA standards. This process involves removing 18 specific patient identifiers and is a combination of automated algorithms and human quality review to balance privacy with data utility for research. - This collaboration is situated within a larger regulatory landscape shaped by the ONC and CMS interoperability rules, which mandate the use of standardized APIs like HL7 FHIR to promote patient data access and prevent information blocking. These regulations are intended to facilitate the secure exchange of electronic health information across different systems and applications. - For a nurse informaticist at a hospital using Epic, a primary challenge will be the integration of AI-driven insights from platforms like Verily's into clinical workflows. Epic supports the integration of third-party AI models and has its own tools for tasks like sepsis detection and clinical summarization. Successful integration often involves leveraging Epic's developer program and adhering to standards like SMART on FHIR. - AI applications in medical imaging, particularly in acute care, focus on enhancing diagnostic accuracy and optimizing workflows. For instance, AI can assist in the early detection of critical conditions like intracranial hemorrhages on CT scans or pulmonary nodules on chest X-rays. This can help prioritize critical cases and reduce the time to diagnosis. - Verily has also been involved in developing clinical decision support tools in partnership with institutions like Mayo Clinic. These tools aim to provide clinicians with concise, relevant, and applicable answers to clinical questions at the point of care, initially focusing on cardiovascular and cardiometabolic conditions. - The rise of AI in healthcare is also leading to new consumer-facing applications, such as Verily's "Me" app, which provides personalized health recommendations from clinicians based on a user's medical records. The app also features an AI companion named "Violet" to answer users' questions about their health data.

Key numbers

The partnership provides researchers with access to Segmed's extensive database of de-identified real-world imaging data, which includes approximately 30 million patients and 150 million imaging exams.
An initial dataset available through the collaboration includes 558 digital breast tomosynthesis exams, with 271 of those having biopsy-proven malignant lesions.
This process involves removing 18 specific patient identifiers and is a combination of automated algorithms and human quality review to balance privacy with data utility for research.
This collaboration is situated within a larger regulatory landscape shaped by the ONC and CMS interoperability rules, which mandate the use of standardized APIs like HL7 FHIR to promote patient data access and prevent information blocking.

What happens next

For a nurse informaticist at a hospital using Epic, a primary challenge will be the integration of AI-driven insights from platforms like Verily's into clinical workflows.
These tools aim to provide clinicians with concise, relevant, and applicable answers to clinical questions at the point of care, initially focusing on cardiovascular and cardiometabolic conditions.
Segmed, a provider of real-world medical data, has partnered with Verily to expand access to imaging data for research.

Sources

Quick answers

What happened in Segmed and Verily Partner on Real-World Imaging Data?

Why does Segmed and Verily Partner on Real-World Imaging Data matter?

Verily's Pre platform is engineered to harmonize and analyze complex biomedical data, empowering researchers to build and deploy AI models. This AI-native platform processes diverse, multimodal data—both structured and unstructured—to accelerate the development of new therapies. A key feature is its ability to transform static PDF protocols into dynamic digital models, which can significantly reduce the time required for study startups. The partnership provides researchers with access to Segmed's extensive database of de-identified real-world imaging data, which includes approximately 30 million patients and 150 million imaging exams. An initial dataset available through the collaboration includes 558 digital breast tomosynthesis exams, with 271 of those having biopsy-proven malignant lesions. To ensure patient privacy, Segmed employs a de-identification process that has received an independent Expert Determination, confirming it meets HIPAA standards. This process involves removing 18 specific patient identifiers and is a combination of automated algorithms and human quality review to balance privacy with data utility for research. This collaboration is situated within a larger regulatory landscape shaped by the ONC and CMS interoperability rules, which mandate the use of standardized APIs like HL7 FHIR to promote patient data access and prevent information blocking. These regulations are intended to facilitate the secure exchange of electronic health information across different systems and applications. For a nurse informaticist at a hospital using Epic, a primary challenge will be the integration of AI-driven insights from platforms like Verily's into clinical workflows. Epic supports the integration of third-party AI models and has its own tools for tasks like sepsis detection and clinical summarization. Successful integration often involves leveraging Epic's developer program and adhering to standards like SMART on FHIR. AI applications in medical imaging, particularly in acute care, focus on enhancing diagnostic accuracy and optimizing workflows. For instance, AI can assist in the early detection of critical conditions like intracranial hemorrhages on CT scans or pulmonary nodules on chest X-rays. This can help prioritize critical cases and reduce the time to diagnosis. Verily has also been involved in developing clinical decision support tools in partnership with institutions like Mayo Clinic. These tools aim to provide clinicians with concise, relevant, and applicable answers to clinical questions at the point of care, initially focusing on cardiovascular and cardiometabolic conditions. The rise of AI in healthcare is also leading to new consumer-facing applications, such as Verily's "Me" app, which provides personalized health recommendations from clinicians based on a user's medical records. The app also features an AI companion named "Violet" to answer users' questions about their health data.