AIToday
Large Language ModelsAI in HealthcareTop Companies' AI MovesTop Companies AIPublished: Nov 15, 2026, 06:30 JST

Thermo Fisher unifies 20 million patient records, launches AI matching on AWS

Send on LINE
Thermo Fisher unifies 20 million patient records, launches AI matching on AWS

3 Key Points

  1. What happened

    Thermo Fisher Scientific built a curated patient recruitment data foundation on AWS, unifying more than 20 million patient profile records (about 1.6 TB), and launched Patient Match AI in production in Q2 FY26.

  2. Why it matters

    Recruitment teams can now run cohort discovery in minutes rather than days, replacing handoffs to technical teams with self-service natural-language queries.

WHO IT HITSClinical trial recruitment teams and site coordinators gain a faster, self-service way to identify eligible patient populations, while IT and data engineering staff may see fewer ad-hoc query requests.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Context & Analysis

Thermo Fisher's patient recruitment data had grown across multiple disconnected systems—patient profiles, surveys, consent records, marketing interactions, and site data—creating fragmentation that made it hard for recruitment teams to get a clear view. Existing processes often required specialized technical teams to translate business questions into database queries, then repeat the cycle as criteria changed. Rather than layering AI on top of these systems, the team first built a modern, curated data foundation on AWS, standardizing and preparing data for analytics and AI workloads. This foundation then supported two complementary capabilities: Central Insights Hub for recruitment analytics and Patient Match AI for natural-language cohort discovery.

A key architectural choice was separating operational application workloads from analytical processing, allowing complex cohort queries to run across large volumes of patient profiles without burdening transactional databases. Thermo Fisher also engaged AWS Countdown Premium from the testing phase onward, which provided a designated engineer who reviewed pipeline design and helped tune Amazon DocumentDB, cutting estimated processing time by more than 3x. Testing at full scale against the complete 20 million record dataset exposed query performance problems early, so they could be fixed before go-live—a factor that likely contributed to a smoother production launch.

The results point to a broader shift in how recruitment teams work: instead of waiting for information to move through multiple technical handoffs, they can interact with trusted data directly, using dashboards to see what is happening and natural-language tools to explore what may be possible. Thermo Fisher says the shared foundation creates an opportunity to continue evolving recruitment capabilities without repeatedly rebuilding the underlying data infrastructure.

FAQ
What is Patient Match AI?
It is Thermo Fisher's proprietary internal tool that supports self-service patient search, feasibility assessment, and cohort discovery through natural-language interaction.
When did Patient Match AI go into production?
It went live in production in Q2 FY26.
How long did AWS Countdown Premium support the launch?
AWS Countdown Premium supported the launch over a four-month engagement.
Top Companies AIRead Original Article
Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleFujikura's 160.6% profit gain leads four wire makers