Triomics has announced the successful raising of $15 million to advance its generative AI-powered platform for oncology. The funding comes from several prominent Silicon Valley investors, including Lightspeed, Nexus Venture Partners, General Catalyst, and Y Combinator. The company’s platform leverages their proprietary generative AI models, OncoLLM™, and specialized software to process unstructured health record data, match patients to clinical trials, and enhance precision oncology.
Triomics addresses a critical issue in oncology where staff currently spend considerable time manually reviewing patient records to find suitable clinical trials or care pathways. This process, which can take hours per patient, often leads to clinical delays and administrative backlogs. By applying generative AI, Triomics aims to automate and expedite these workflows.
Co-founders Sarim Khan and Hrituraj Singh, who have backgrounds in biotech research and AI, recognized the potential of recent AI advancements to analyze the vast amounts of unstructured medical data. Their collaboration with the Medical College of Wisconsin led to the development of OncoLLM™, which can identify eligible patients for clinical trials in minutes, significantly faster than traditional methods. The model also demonstrates high accuracy in extracting data from unstructured notes and performs cost-effectively compared to existing models like GPT-4.
Triomics has also published results indicating that their information retrieval engine for oncology outperforms other advanced retrieval models by 1.5 to 2 times. Their proprietary software, such as Triomics Prism and Triomics Harmony, integrates with electronic health records (EHRs) to match patients with clinical trials and support quality reporting and precision oncology.
Bradley Taylor of the Medical College of Wisconsin praised Triomics for its collaborative approach and evidence-based development of their models. Anai Kothari, a surgical oncologist at the same institution, highlighted the importance of Triomics’ ability to convert complex cancer data into actionable formats to improve patient care.
The company’s future plans include publishing additional efficacy data for OncoLLM™ across diverse settings and developing software to support more use cases. Triomics partners with leading academic cancer centers and organizations such as the Collaboration for Oncology-focused LLM Training (COLT) and the Cancer Informatics for Cancer Centers (CI4CC) Society to establish performance and safety benchmarks.
Triomics co-founders Sarim Khan (L) and Hrituraj Singh (R)
