Editors: Raymond D. Aller, MD, & Dennis Winsten
ARUP grant supports lab innovation collaboration
September 2024—The ARUP Institute for Research and Innovation in Diagnostic and Precision Medicine has created the Sherrie Perkins Research and Innovation Collaboration grant to fund lab medicine research that has the potential to significantly improve patient care.
The grant is designed to foster collaboration within the medical community and may award up to $150,000 to industry and academic partners, who will work with ARUP medical directors and scientists.
“We hope to kickstart collaborations between ARUP Laboratories and other industry and academic partners, especially smaller companies that may not have as many resources,” said Leo Lin, MD, PhD, an ARUP medical director of immunology, PharmaDx, and the research and innovation institute.
The grant is the latest offering from the institute, which was founded one year ago. The grant is named after retired ARUP CEO Sherrie Perkins, MD, PhD, who “laid the groundwork for an increased focus on innovation at ARUP,” according to a press release from the national reference lab, a nonprofit enterprise of the University of Utah.
Paige releases open-source AI pathology models for research
The artificial intelligence-driven pathology solutions provider Paige has announced the open-source release of Open Paige Foundation Model, which includes Virchow and Pathology Report and Image Summarization Model, both of which were developed through a collaboration with Microsoft Research.
Virchow, a foundation model for computational pathology, is based on a globally diverse data set of 1.5 million digitized pathology slides from Memorial Sloan Kettering Cancer Center. PRISM, which is built on Virchow, is a vision and language model that can automatically generate diagnostic summaries from H&E-stained whole slide images for research purposes.
The models are available on the Hugging Face machine learning and data science collaboration platform and community, allowing AI developers, academic researchers, and innovators to create customized AI models and applications that support cancer research and diagnostics.
More advanced versions of Virchow, PRISM, and other recent AI models from Paige are accessible through commercial licensing.
Regenstrief Institute announces LOINC semiannual update
The Regenstrief Institute has released a semiannual content update for the Logical Observation Identifiers Names and Codes, its global standard for health terminology.
LOINC release 2.78 is intended to bolster the electronic exchange of laboratory and clinical data through such measures as adding nearly 1,600 new concepts and modifying approximately 1,400 existing concepts. The content focuses on such areas as diagnostics, medical devices, public health reporting, and social determinants of health. The new release also includes nearly 250 new concepts to standardize the representation of data for semantic interoperability.
Regenstrief’s Health Data Standards unit, in collaboration with stakeholders, has created new concepts for an FDA-approved diagnostic system that uses artificial intelligence to identify cervical cancer and concepts to standardize data from continuous glucose monitoring devices, developed in partnership with the Diabetes Technology Society. It too has collaborated with the Department of Health in Australasia on concepts to standardize pathology data in laboratory information systems.
LOINC is available royalty free. Regenstrief releases semiannual updates in February and August.
For further information about what is offered in version 2.78 or to submit queries or collaborate with others in the LOINC community, visit forum.loinc.org.
Indica Labs launches tool for macrodissection workflows
Indica Labs has introduced Lung Macrodissect AI, its artificial intelligence tool for streamlining macrodissection workflows.
“Lung Macrodissect AI quantifies and standardizes tumor content assessment, reducing subjectivity and errors while streamlining the macrodissection annotation process for primary and metastatic non-small cell lung cancers,” according to a company press release.
The tool is integrated in Indica’s Halo AP enterprise digital pathology platform and can be combined with Xyall’s Tissector line of robotic macrodissectors for a fully automated, audited workflow. It too can be used to guide and standardize manual macrodissection workflows.
Beyond highlighting cancerous tissue, Lung Macrodissect AI calculates the number of tumor cells in an annotated region of interest and the percentage of tumor present in all annotated regions of interest, ensuring that only adequately concentrated specimens are sent for downstream genetic sequencing and molecular testing.
Lung Macrodissect AI is available for research use only in the United States.
Indica Labs, 505-492-0979