ボストン小児病院が AI を活用して新たな診断を可能に

Boston Children’s Hospital did not pursue artificial intelligence simply to experiment with new technology. The hospital embedded AI across the organization as a core part of its clinical and operational infrastructure to improve how care is delivered to its pediatric patients, particularly those with complex and rare conditions. By integrating AI into daily workflows, the team has reduced operational costs, improved access to care, and helped diagnose more than 40 rare conditions that had previously gone unresolved.

Operating under pressure

Boston Children’s Hospital is one of the largest pediatric institutions in the world, serving patients across more than 40 specialties with close to 1 million outpatient visits each year.

Like many health systems, it operates under tight financial constraints while managing increasing administrative burden. Teams across supply chain, billing and operations handle high volumes of repetitive tasks, from processing invoices to coordinating schedules. These processes are necessary but time-intensive, pulling staff away from higher-value work.

At the same time, clinical teams face a different kind of limitation. Rare disease cases often involve fragmented genetic data, incomplete clinical histories and an overwhelming body of medical literature. Even in a leading research institution, physicians cannot synthesize all of that information fast enough to reach every diagnosis.

“The problem isn’t effort,” says John Brownstein, Chief Innovation Officer at Boston Children’s. “It’s human cognitive limits.”

Setting the foundation with an enterprise AI layer

Boston Children’s began with individual AI use cases, including documentation and translation tools. But those early efforts quickly exposed the limits of a fragmented approach.

“You cannot just rely on one-off solutions,” Brownstein says.

The hospital shifted to building what Brownstein calls an enterprise AI layer: a secure internal ChatGPT environment used across research, clinical, and administrative teams. Instead of treating AI as a collection of tools, the organization created a shared foundation where new capabilities could be developed and deployed quickly.

This system allows teams to work with AI in ways that are directly relevant to their roles, whether that involves accessing internal data, synthesizing medical literature or streamlining workflows. Governance structures were built alongside the technology to ensure safety, monitoring and consistent evaluation.

The shift changed the pace of innovation. Tools that once required extended development cycles can now be deployed in days, allowing the organization to respond quickly to both operational demands and clinical needs.

Today, more than one-third of employees use AI as part of their daily work, spanning clinical, research and administrative functions.

Redesigning workflows across operations

Boston Children’s focused first on areas where AI could deliver measurable operational impact. In supply chain operations, AI now manages invoice intake, routing and responses.

In parallel, the hospital applied AI to surgical scheduling. By analyzing clinical notes and estimating patient acuity, the system improves how operating room time is allocated. This allows schedules to be planned further in advance, increasing utilization and enabling more patients to receive the care that they need faster.

Additionally, physicians use AI for decision support and to synthesize complex clinical information. Researchers apply it to data analysis and cohort building. Administrative teams rely on it for drafting documents, coding and improving workflows.

The organization ties these changes directly to measurable outcomes. Across more than 50 automations, Boston Children’s has captured about 60,000 hours in time savings, which is equivalent to more than $7 million in redeployed labor.

The organization has focused on making AI relevant to everyday work rather than introducing it as a standalone initiative.

“The key here is meeting people where they are,” Brownstein says.

Advancing rare disease diagnosis and genetic research

Alongside operational improvements, Boston Children’s invested in AI for clinical discovery. The hospital developed what it describes as a “co-pilot geneticist,” designed to integrate genetic data, phenotypic information and global medical literature.

This system addresses one of the most difficult challenges in medicine: diagnosing rare diseases that have eluded explanation for years.

As a result of this work, more than 40 diagnoses have been made to date that were previously thought impossible. The work has also led to the identification of new gene targets and potential therapeutic pathways.

“We combine genetic information, phenotypic information, literature search, and the reasoning of AI to deliver diagnoses to families that were once left without any answers,” Brownstein says.

For patients and families, the impact is immediate and tangible. Cases that once remained unresolved are now yielding answers and, in some instances, new directions for treatment.

“This was unthinkable before, but is now providing hope to so many families,” Brownstein says.

AI-enabled care at scale

The next phase of Boston Children’s AI strategy focuses on deeper integration and broader adoption. Leadership sees significant opportunity to expand both usage and impact.

The hospital is working to embed AI more fully into clinical decision-making, extend tools across specialties and continue refining models through collaboration with OpenAI.

Over time, AI is expected to become a core component of medical practice.

“How would you not want an incredibly trained physician alongside all the world’s medical knowledge?” Brownstein said.

At Boston Children’s, AI is becoming part of the infrastructure that supports care delivery, research and discovery—redefining what’s possible for both clinicians and patients.