Back to Projects

CASE STUDY

TrialForge AI

AI-assisted clinical trial documentation and protocol drafting, combining biostatistics expertise with OpenAI-powered generation to accelerate trial setup.

PythonOpenAI APIFastAPIClinical Trials

AI-Drafted

Protocol documentation generated from structured inputs

Validated

Cross-checked against trial design consistency rules

Biostatistics-Led

Prompt engineering grounded in trial methodology expertise

The Problem

Drafting clinical trial protocol documentation is slow, repetitive, and error-prone — coordinators spend significant time writing sections that follow predictable structures, while inconsistencies between sections can delay regulatory approval.

The Solution

I built an AI-assisted documentation tool that drafts protocol sections from structured trial parameters, checks them for internal consistency, and gives coordinators a structured review workflow before finalising submission-ready documents.

AI-Assisted Documentation

Generates and structures clinical trial protocol documents using OpenAI API-driven drafting.

Protocol Consistency Checks

Cross-references drafted sections against trial design rules to flag contradictions early.

Structured Workflow Engine

Guides trial coordinators through a repeatable, auditable documentation process.

Biostatistics-Informed Prompts

Prompt design grounded in biostatistics methodology to reduce hallucinated or invalid content.

How It Works

01

Trial Parameters Input

Coordinator enters trial design parameters — population, endpoints, arms, and methodology.

02

AI Drafting

OpenAI API drafts protocol sections and supporting documentation based on the input parameters.

03

Consistency Validation

Drafted content is checked against trial design rules for contradictions or gaps.

04

Human Review

Coordinator reviews and edits AI-drafted sections before finalising.

05

Export & Archive

Finalised protocol documentation is exported and archived for regulatory submission.

Outcome

TrialForge AI cuts the time required to produce first-draft protocol documentation, while its consistency checks catch design contradictions earlier in the process, reducing costly revisions later in regulatory review.

Back to all Projects