Keiji AI
Research Flow
The Research flow is a conversation, where you can ask Keiji AI to conduct a literature review, search public databases for clinical trials, and form courses of study to conduct research.
By having access to rich, curated clinical trial and biomedical data sources, TrialMind can help you plan your studies while referencing the state of the art in your field.
Examples
Data Science Flow
The data science flow is a user experience focused on collaboratively generating code to fulfill a task. TrialMind's code generation is catered to your team and utilizes your historical code base to generate code that is consistent with your team's style and conventions.
How it works
The DS flow has 2 stages.
Stage 1: Index Historical Code
The user provides a historical code base to TrialMind. TrialMind indexes the code base and learns the patterns of the code base, including commonly used processes, and standards.
The user can add to the historical code base at any time, and TrialMind will re-index the code base to learn the new patterns. There is no restriction to how small or how large the code base which should be indexed.
Stage 2: Generate Code
The user provides a task to TrialMind. TrialMind generates code to fulfill the task. The user can provide feedback to TrialMind to improve the code generation. TrialMind will learn from the feedback and improve the code generation.
The generated code will contain structures, patterns and conventions which your team uses. This generated model will be consistent with any propriatery data models your team utilizes. Because the generated code is consistent with your projects, it can be quickly integrated into your code base.
Examples
Demo on Patient Event data for a clinical trial.