Experts show how AI tools boost diabetes education and EHR efficiency
WASHINGTON D.C., UNITED STATES — Artificial intelligence is reshaping healthcare by tackling two critical challenges: empowering young diabetes patients and streamlining physician access to complex medical data.
At the AWS Summit Washington, DC 2025, experts showcased how AI tools enhance patient education and clinical decision-making, demonstrating that AI’s impact extends far beyond ambient documentation, as reported by Jordan Scott in HealthTech Magazine.
Generative AI personalizes diabetes education for kids
Steven Silvers, a Harvard research assistant and game developer, created T1D Learning Camp to educate children through generative AI-powered interactive conversations.
The game adapts in real time, reinforcing lessons on nutrition and blood sugar management while allowing children to discuss their experiences with AI-powered characters.
The game leverages Amazon Web Services’ AI suite, including Amazon Bedrock for generative responses, Amazon Polly for speech synthesis, and Amazon Transcribe for voice recognition. It also utilizes Amazon Titan for culturally tailored food imagery, ensuring relatability across diverse dietary preferences.
By merging education with play, the game provides a scalable and engaging solution, demonstrating that AI can make chronic disease management more accessible and effective for young patients.
Hybrid AI search unlocks EHR insights for physicians
Clinicians often struggle to extract meaningful insights from fragmented electronic health records (EHRs), particularly in cases involving complex conditions.
Dr. Dinesh Rai, clinical AI engineer at the Innovation and Digital Health Accelerator at Boston Children’s Hospital, developed an AI-powered hybrid search engine that combs through structured and unstructured data, transforming vague hunches into data-driven decisions.
The tool automates patient cohort creation, semantic tagging, and vector-based searches, allowing doctors to retrieve historical precedents in seconds.
The system is a natural-language query system built on AWS infrastructure that transforms queries into searchable objects, searching structured query language (SQL) databases and vectorized records.
“When a user inputs a query, the agent takes that natural-language query and transforms it into an object to search the SQL database,” Dr. Rai said.
In closing loopholes in the applicability of EHRs, the ingenuity of Rai depicts how AI can demystify clinical workflow and transform bedside care into personalized practice.
