Team

‍Our History
‍Since our founding, OmniScience has partnered with over 50 life sciences teams, applying advanced AI and machine learning to hundreds of clinical trial datasets to support precision diagnostics, biomarker analytics, clinical trial analytics, clinical decision support, and market intelligence. Over the past five years, we’ve evolved from bespoke AI development to productizing our innovations, culminating in our flagship SaaS platform: Vivo.

Translating complex clinical trial data into knowledge
Our team has applied every AI/ML technique to every kind of clinical trial data - to create knowledge to support the development of many solutions.

• We have analyzed wearables and DHTs to identify digital biomarkers of behavior from video, voice, ECG, EEG, and actigraphy data for complex CNS studies. Read more here and here.
• We have applied machine learning to genomics and proteomics datasets to support precision diagnostics and drug development. Read more here and here.
• We have modeled participant engagement to identify clinical trial participant dropout risk. Read more.
• We have analyzed large clinical trial datasets to support clinical trial planning and feasibility.
• We have conducted bespoke modeling of clinical trial data including novel medical devices to support oncology diagnoses and treatment decisions.  Read more here and here.
• We have applied computer vision to radiology and pathology images to develop clinical decision support tools.  Read more here.
• We have applied natural language processing (NLP) and genAI to many kinds of unstructured data to extract insights, sentiment, trends, scientific concepts, gene-disease-drug relationships, market intelligence, and more. Read more here, here, here, and here.
• We have automated clinical trial participant data processing to enable participant stratification, endpoint prediction, and biomarker discovery. Read more here.

‍

A decade ahead in language models
Our team’s work with NLP started in their PhD training in the early collaboration with IBM Watson, Baylor College of Medicine, and MD Anderson Cancer Center to mine scientific literature to support treatment decisions. ‍Before large language models (LLMs) entered the mainstream, we were already applying NLP to the hardest problems in biomedical data, extracting insights from fragmented datasets, synthesizing unstructured scientific content, and delivering language-driven biomarker analytics and decision support tools. Our early NLP work laid the foundation for the genAI era.

From analytics to actionable insights
‍With the advent of LLMs, we accelerated our innovation. By tuning language models with biomedical vocabularies and ontologies, combined with our expertise in working with complex clinical trial data, we created domain-specific solutions that are interpretable, robust, and impactful.

These capabilities are deeply integrated and grounded in scientific context, enabling better, faster decisions across clinical operations and strategy.

For more details about how we’re applying LLMs and genAI in our products, see our genAI whitepaper →