platoseedplatoseed
AI Agents for Oncology EHRs
Triomics is building the AI infrastructure for modern oncology. Across the U.S., cancer care providers rely on highly trained staff to manually review unstructured patient records such as pathology reports, clinical notes, genomic panels, and scanned faxes to support workflows like clinical trial matching, visit preparation, and quality reporting. We replace that manual work with task-driven AI agents embedded directly into clinical workflows, automatically processing medical records at scale and in real time. Our platform is trusted by 4 of the top 10 Best Hospitals for Cancer ranked by U.S. News, as well as several of the largest community oncology practices.
Triomics provides AI agents that read across oncology patient records, transforming unstructured EHR data into structured, citable insights usable within cancer care workflows. Their platform aims to enable pre-charting, trial matching, visit preparation, and registry curation to support oncology centers.
The product uses AI agents to reason across entire patient records (notes, reports, faxes, pathology, molecular data, imaging) and deliver structured, cited answers directly into oncology workflows. Key capabilities include patient trial matching with a ready-to-use, cited worklist, pre-charting for visit preparation with disease status and biomarkers, and registry curation with abstraction across NAACCR/SEER/COC/QOPI measures. Outputs are traceable to specific lines in the chart, with source citations and audit trails embedded.
Who it’s for: Cancer centers and oncology departments, including large cancer centers and community oncology practices, that require structured patient views, trial matching, visit preparation, and registry abstraction.
Latest funding mention ($22M raise) and customer references; multiple case studies and abstracts indicating traction with notable cancer centers.
I am a chemical engineer with previous research experience in tissue engineering and neuroscience. The lack of vertical integration in the clinical trial space inspired me to work on Triomics.
Before starting Triomics, I was a Researcher at Adobe Research, where I worked on language models and reinforcement learning problems. I now use the same technologies to interpret oncology patient charts.
Analyzing EHR and messaging data to predict disease before it occurs.
Software for remote monitoring of cancer patients
