![]()
U.S. Patent No. 12,783,788 covers turning faxes, PDFs, and scanned records into decision-ready documents before they reach the person who has to act on them.
OAKLAND, CA, UNITED STATES, November 4, 2026 /EINPresswire.com/ — xCures, the clinical clarity engine for healthcare, today announced that the United States Patent and Trademark Office has issued U.S. Patent No. 12,783,788, “Dynamic In-Transit Structuring of Unstructured Medical Documents.” The patent, the company’s second, covers the method at the center of the xCures Clinical Clarity Engine.
Healthcare can move a patient’s records across the country in minutes, and the recipient still cannot read them. A single records request to a national network returns duplicates, partial responses, PDFs, faxes, and scans, usually as one bundle that a person has to sort by hand. The patented method sorts the bundle while it is in transit. The system parses the bundle into its component sub-documents and classifies each one, whether pathology, imaging, genomics, labs, clinic and progress notes, or diagnostic and prognostic reports. Disease-specific ontologies drive the metadata extraction, and a manifest, in effect a table of contents, indexes the whole package. The recipient gets the structured result in whatever format their system takes. Nothing is left to sort.
“The best of medicine should belong to everyone, and today it belongs to whoever can afford the staff to read the chart,” said Mika Newton, Chief Executive Officer of xCures. “xCures exists to give healthcare the capacity to care for everyone. This patent protects the method that creates that capacity. Records move freely now, and most of what arrives is still unreadable at the point of decision. We structure the record in transit and link every element back to its source document, so a reviewer can check our work against the original. A doctor makes a better decision with it, and an organization serves more people with fewer errors and lower costs.”
The method runs today across the national networks xCures uses to retrieve records: Carequality, eHealth Exchange, and TEFCA through Kno2 as a Qualified Health Information Network. The engine processes more than 1.5 million documents a day from more than 550,000 care locations across all 50 states. The structured output feeds xCures Decision-Ready Checklists, the company’s flagship extraction product. A peer-reviewed study published this year in AI in Precision Oncology reported checklist accuracy of 97.0% to 98.2%.
“Most of the clinical record still travels as unstructured documents, and that will stay true for years,” said Glenn Kramer, PhD, Chief Technology Officer of xCures and a named inventor. “We decided early that the structuring had to happen while the document is in motion, at the volume the networks actually return, with a manifest the recipient can trust. The classifier improves with every bundle it sees. The claims cover that architecture end to end, from parsing and classification through ontology-driven metadata, packaging, and delivery. Customers never see this layer, and everything else we build runs on top of it.”
“Every patient in this system has a history scattered across the places they received care,” said Mark Shapiro, lead inventor on the paten. “We built this method because we were drowning in the paper that comes back when you ask for a complete record. A 200-page fax might hold several MRI reads, a pathology report, a genomics report, and a stack of clinic notes. A clinician starts by working out which is which and what each one says about the disease. We automated that first step so the reasoning can start sooner.”
“What lands in a customer’s system is a pathology report, a genomics report, and a progress note, each labeled and indexed, in the format their system takes,” said Bryan Federowicz, Vice President of Sales at xCures and a named inventor. “A lab routes each report to the right workflow the moment it arrives, and a clinician opens the one document that matters without scrolling through the rest. Our customers buy that outcome, and this patent protects the step that produces it.”
The patent names Mark A. Shapiro, Bryan J. Federowicz, and Glenn A. Kramer as inventors. Wilson Sonsini Goodrich & Rosati prosecuted the application. It issues with 50 claims and a patent term adjustment of 807 days. xCures has filed a continuation application extending the family to system and computer-readable-medium claims, and additional applications remain pending in the United States and China. The company’s first issued U.S. patent, No. 11,887,738, “Platforms for Conducting Virtual Trials,” was granted in January 2024.
The xCures Engine is HITRUST i1 certified. Its customers span diagnostics, telehealth and digital health, health systems, value-based care, and specialty pharmacy, and include Caris, Veracyte, NeoGenomics, and Exact Sciences.
About xCures
xCures® is the Clinical Clarity Engine for healthcare, retrieving medical records and transforming them into decision-ready data. The company is focused on improving healthcare outcomes by increasing the value of medical records. The Engine is built on almost a decade of work in the most complex area of medicine, oncology, and is now used across all clinical domains, scaling across care and workflows.
The Engine extracts decision-ready data from medical records retrieved across U.S. healthcare systems and delivers it via a web UI or a developer-friendly API. Every answer links to its source document, enabling teams to make better-informed decisions.
For more information, contact info@xcures.com or visit xcures.com.
xCures® is a registered trademark of xCures, Inc.
Patrick van der Valk
xCures Inc
Visit us on social media:
LinkedIn
The xCures Clinical Clarity Engine
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
![]()
Media gallery
