Triomics secures $15M to boost oncology workflows with AI
Triomics, a health technology firm, has announced a successful $15 million Series A funding round aimed at enhancing the efficiency of oncology workflows in the United States. The company utilises advanced generative AI models and specialised software to process free-text health records, match patients to clinical trials, and improve the overall quality of cancer care.
Current oncology staff face the laborious task of manually combing through thousands of patient health records to identify suitable clinical trials or care pathways. Triomics aims to alleviate this burden by automating these processes with its generative AI-powered platform, known as OncoLLM. Investors in this funding round include notable Silicon Valley firms such as Lightspeed, Nexus Venture Partners, General Catalyst, and Y Combinator.
Manual review of patient records can take hours per individual, leading to significant delays in clinical workflows and impacting patient care. These delays can result in patients missing out on clinical trials or specialised treatments, and can contribute to provider dissatisfaction and staff turnover. Triomics co-founders Sarim Khan and Hrituraj Singh, both experts in AI and biotechnology, have developed solutions to address these challenges.
"Hrituraj and I decided to partner to build solutions leveraging the advances in the field of generative AI and LLMs to help hospital staff," said Sarim Khan, CEO of Triomics. "We want our solutions to reason and sound like experts in oncology."
In collaboration with researchers from the Medical College of Wisconsin, Triomics has made significant strides. Their generative AI model, OncoLLM, demonstrated the ability to identify 90% of eligible patients for clinical trials within minutes—a task that typically takes days or weeks for qualified nurses. The model also extracted structured data points from unstructured notes with accuracy comparable to or exceeding other established AI models while being substantially more cost-effective.
"Most of the solutions on the market today claim they use generative AI, but many lack published evidence. Triomics is setting themselves apart by taking a truly collaborative approach to co-developing these models," said Bradley Taylor, Chief Research Informatics Officer at the Medical College of Wisconsin.
Triomics' proprietary software integrates with existing health system electronic health records (EHRs) to complete clinical and administrative tasks. For instance, Triomics Prism aids in matching patients to clinical trials by pre-screening oncology patients with upcoming appointments. Triomics Harmony helps curate EHR data to support quality reporting and precision oncology.
"Our investments have been deliberate. We have merged expertise in AI with clinical experience in oncology to complement the strengths of our advanced models while addressing potential flaws," commented Hrituraj Singh, CTO of Triomics.
The company also partners with leading academic cancer centres and researchers to set benchmarks and best practices for generative AI performance and safety. Partners include the Collaboration for Oncology-focused LLM Training (COLT) and the Cancer Informatics for Cancer Centers (CI4CC) Society.
"We differentiate ourselves by building tailored models specifically for oncology and pairing them with generative AI workflows," said Sarim Khan. "Other solutions use broad-based technologies that lack the scalability and return on investment needed by the industry."
Triomics plans to continue publishing data on OncoLLM's effectiveness across diverse settings and patient populations. The company will also develop additional software to power more use cases, furthering its mission to automate clinical trials and streamline oncology workflows.
"With robust early results for their proprietary oncology-specific models and partnerships with leading cancer care centres, Triomics is well poised to deliver significant value to cancer care providers and patients," said Jishnu Bhattacharjee, managing director at Nexus Venture Partners.