About TrialBridge
Better-organised, more transparent research operations.
Aspirox TrialBridge is a product in development by Core Aspirox. It is built around one idea: every step from protocol to site should be traceable to the step before it.
Help research teams connect the right patients to the right trials — with every decision on the record.
Recruitment depends on many careful judgements by coordinators, reviewers and investigators. We want to give that work better structure: criteria that cite their source, data gaps that stay visible, queues that everyone shares, and numbers that can be checked.
The mark — two posts joined by a span, with a single node near one end — stands for the protocol, the research site, the criteria that connect them, and the candidate in between.
Six principles we hold the product to.
Evidence over assertion
Every structured item cites where it came from — a protocol passage, a dataset field, a person.
Unknown is a first-class value
Missing information is shown as Unknown. It is never quietly turned into a “no”.
People decide
Software proposes, filters and organises. Qualified people approve criteria and decide outcomes.
Versions are immutable
Approvals, cohort runs and decisions pin to the exact criteria version they used.
Least data necessary
Pseudonymous candidates by default; identity stays with the organisation that holds it.
Every number is traceable
A metric links to the records it counts, with its calculation written down.
Fit the work research teams already do.
TrialBridge isn’t a replacement for a CTMS, EDC or consent system, and it doesn’t replace investigators. It sits in the part of the workflow that is often run on spreadsheets: turning a protocol into a checklist, checking it against data, and coordinating who reviews whom.
We start from the roles people already have — coordinators, reviewers, data managers, investigators — and give each one a clear view scoped to their work.
A narrow, checkable role for AI.
We use AI for one task: proposing structured criteria from protocol text, each one citing a passage we can verify. It never sees patient data, never approves anything and never screens a candidate.
We’ll evaluate that feature against expert-annotated protocols before it ships, and we won’t publish accuracy claims until that evaluation exists. More on AI boundaries.
What we will and won’t say.
- We label every product preview that uses synthetic data, and every capability that is planned rather than built.
- We don’t show customer logos, testimonials, partnerships or results we don’t have.
- We don’t claim certifications or regulatory status until they are independently established.
- When something on this site doesn’t work yet — like sign-in or the contact form’s delivery — we say so.
Help shape the first release.
We’re looking for research teams willing to tell us where their recruitment workflow hurts. Start with a conversation.