PicnicHealth unveils PicnicAI to glean real-world data from medical records with advanced generative AI

PicnicHealth‘s PicnicAI uses advanced AI to aggregate and interpret US electronic health records (EHRs), improving personalized care and research efficiency with enhanced accuracy and universal access to patient data.

PicnicHealth has introduced PicnicAI, a technology designed to unlock the Universal Patient Record. This platform gathers EHRs from across the US and converts them into valuable data for both patients and researchers.

At the core of PicnicAI is a proprietary, industry-leading large language model (LLM) that outperforms GPT-4 and other healthcare-specific models in accuracy, efficiency, and cost when structuring and interpreting medical records. The LLM, trained on over 350 million annotations from records spanning more than 100,000 care sites, accurately organizes and abstracts comprehensive patient records into actionable insights.

PicnicAI serves as the foundation for PicnicHealth’s patient and life sciences products. For patients, it enables personalized care based on individual medical histories and needs. In the life sciences sector, PicnicAI streamlines observational research, allowing companies to achieve study endpoints more rapidly and adopt advanced technology responsibly.

Key features and benefits:

• Universal access: PicnicAI connects to 100% of US care sites, overcoming fragmented or inaccessible patient data issues.
• Rapid retrieval: After patient enrollment, PicnicAI provides instant access to most records and over 90% of key records within 2 weeks.
• Bias reduction: The diverse training data helps minimize biases related to treatment patterns, race, ethnicity, or socioeconomics.
• Superior performance: In evaluations, PicnicAI’s LLM performed three-times better than GPT-4 for clinical named entity recognition, and 47% better than leading medical LLMs in record abstraction.
• Cost efficiency: PicnicAI demonstrated a 90% increase in cost-efficiency per record compared to other models.

“Personalized medicine requires technology like PicnicAI,” remarked Troy Astorino, CTO and co-founder of PicnicHealth. “It has taken us a decade to build, but the results feel magical. We can now instantly construct and understand any patient’s complete care journey, pulling records together from every hospital and doctor’s office where that patient has been seen. This allows us to provide clinical insights to patients and our research partners with speed, efficiency, and scale not possible anywhere else. It’s how patient records should work.”

PicnicHealth plans to continue improving PicnicAI and maximizing the capabilities of its advanced medical LLM and platform. The company aims to use the insights provided by PicnicAI to advance the development of effective therapeutics and personalized medical care. In a related discussion on health data management, Nancy A Dreyer, Chief Scientific Advisor, PicnicHealth speaks with Patti Peeples in an episode of ‘Patti’s People.’

---

You may also be interested in:

Patti’s People – Patti Peeples speaks with Nancy A. Dreyer