Dallas-based Children's Health has deployed a generative AI solution from Pieces Technologies Inc. and has become an investor in the company.
Pieces has completed a $25 million financing round with participation from Children's Health, Concord Health Partners, OSF HealthCare, and Rittenhouse Ventures.
Dallas-based Pieces specializes in areas such as applied clinical generative AI, predictive modeling, and physician-supervised machine learning to streamline clinician workflows and improve patient, financial, and operational outcomes.
The company says it optimizes clinical workflows and reduces provider burden for health systems by producing autonomous, AI-generated clinical documentation for multi-disciplinary care teams. Pieces says it has generated more than 5.4 million inpatient clinical summaries across multiple health system clients.
"We are honored to receive support from major health systems and leading healthcare investors at such an exciting inflection point ," said Ruben Amarasingham, M.D., CEO of Pieces, in a statement. "Our mission is to help those who help others. We are grateful that our investing partners recognize our work to help doctors, nurses and care managers free up valuable time so they can stay focused on care delivery at the bedside."
Children’s Health has deployed Pieces’ generative AI solution across its inpatient clinical enterprise. The solution is integrated into the pediatric health system's Epic EHR and used by doctors, nurses and case managers for clinical handoffs, length-of-stay management, discharge barrier resolution, team communication and multidisciplinary rounds.
"We are encouraged by the positive responses to this technology from our care teams," said Philip Bernard, M.D., chief medical information officer at Children's Health, in a statement. "Our clinicians recognize that generative AI will allow our teams to enhance high-quality, innovative care. Pieces is delivering meaningful and generative AI for our patients, ensuring the safe use of this technology today."
The Pieces "working summary" is a 100-word distillation of a patient's course in the health system, despite the length of stay or complexity of care. The summary is generated in real-time by Pieces across a range of clinical disciplines as diverse as hospital medicine, child psychiatry, surgical subspecialties, multi-specialty ICUs and neonatal medicine. Pieces employs a combination of contextual modeling, collaborative and adversarial AI, human-in-the-loop supervision and automated clinical prompt engineering to achieve an extremely low rate of serious error, the company said.
One challenge for AI systems operating in pediatric environments is adequately accounting for the differences in clinical presentation, physiology and treatment from widely varying age groups. In this deployment, Pieces was able to demonstrate successful clinical summarization across age ranges after extensive training and iteration, the company said.
