Boston Children's uses OpenAI AI to diagnose rare diseases

Boston Children's Hospital — one of the largest pediatric medical centers in the USA — has deployed an AI system based on OpenAI's technologies to diagnose rare diseases. Over its operational period, the system has helped identify more than 40 rare diseases and significantly reduced the workload on medical staff.

Why Rare Diseases Are the Main Problem

Rare genetic and hereditary diseases affect a small portion of the population, but they are difficult to diagnose. A child may visit multiple doctors for years, undergo useless treatments, until someone notices an unusual symptom pattern. In the USA, a disease is considered rare if fewer than 200,000 people suffer from it — that's thousands of different diagnoses.

How the AI Assistant Works

Boston Children's integrated a GPT-based system that analyzes the patient's clinical picture: symptoms, test results, disease history. The doctor enters the data — the AI within seconds compares it with a knowledge base of rare diseases and suggests several diagnostic hypotheses with explanations of the logic. The system does not make the decision — the doctor does. But it accelerates the first stage: searching for a rare disease among thousands of possibilities.

• Real-time analysis of clinical data
• Matching symptoms against a database of rare diseases
• Explanation: why the system proposed exactly this hypothesis
• Integration with electronic medical records

Result: 40+ Diagnosed Cases

For the hospital, this is a historic improvement. More than 40 rare diseases that previously could have remained unrecognized for years are now diagnosed within weeks or months. For each family, this means precise treatment instead of symptomatic care. Most importantly: many of these children live in other parts of the USA and came to Boston Children's by referral. AI has not only improved diagnosis within the hospital but made it a center of attraction for the most complex cases.

"We see the result in families' lives.

When we finally find a rare disease, parents get answers and can begin the right treatment."

Reducing Doctor Workload, Better Care

Another effect is operational. Previously, a rare disease specialist could spend hours in medical literature, consulting with specialists from other hospitals, re-taking patient history. Now the initial analysis is AI's job: it will propose 5-10 hypotheses with explanations in minutes. The doctor checks, clarifies, sometimes requests additional tests. His work becomes deeper: not searching for a needle in a haystack, but validating a hypothesis. This is not only faster but more accurate.

What This Means

For rare disease medicine, this is a prototype of the future: AI as a doctor's assistant, not a replacement. The system narrows the search field, the doctor makes the diagnosis. For US Healthcare, this could mean reducing diagnostic delays and decreasing unnecessary procedures.