Stratum Research

Applied LLM Systems for Clinical Reasoning

What We're Building

RAGkit

RAGkit is a JSON first, research-driven toolkit designed to let researchers easily spin up RAG systems and establish standards for Retrieval-Augmented Generation (RAG) systems in academic and industrial research.

GitHub PyPackage

OrthoQA-300

✨0 + Total downloads on HuggingFace

OrthoQA-300 is a structured, synthetic dataset of 300 patient-provider style question-and-answer (QA) pairs focused on orthopedic surgery. Each entry simulates a realistic clinical interaction, with patient-style questions and LLM-generated provider-style answers. Questions are grouped by procedure (e.g., ACL Reconstruction, Total Hip Replacement) and theme (e.g., "What is it?", "Recovery", "Risks").

GitHub HuggingFace

OrthoQA-1k

Building on OrthoQA-300, this synthetic dataset is an expansion with 1000 patient-provider style question-and-answer (QA) pairs focused on orthopedic surgery!

Github HuggingFace

DermQA-1k

DermQA-1k is a synthetic dataset of 1000 patient-provider style question-and-answer (QA) pairs focused on dermatology. Each entry simulates a realistic clinical interaction, with patient-style questions and LLM-generated provider-style answers. Questions are grouped by procedure (e.g., "What is it?", "Recovery", "Risks").

Github