Our Mission
RadOncAI is designed to help Radiation Oncology residents prepare for their oral board examinations. By simulating realistic case discussions, we provide a safe, low-stakes environment for trainees to practice their diagnostic reasoning and management planning.
How It Works
This application utilizes a Retrieval-Augmented Generation (RAG) architecture. Unlike standard chatbots, our system retrieves context from verified Radiation Oncology guidelines before generating an examiner response.
- Frontend: AngularJS & Bootstrap
- Backend: Node.js & Express
- AI Model: OpenAI OSS, Llama 3.2 70B
Under the Hood
When you submit an answer, the system analyzes your response against NCCN guidelines and generates a follow-up question tailored to your specific management plan, mimicking a real examiner's adaptability.
The Team
Seonghoo (Paul) Kim
MS Machine Learning and Data Science Student
Northwestern University
Former Quality Development Engineer at InterSystems with a passion for healthcare technology.
Teo Troy Peng
Radiation Oncology Instructor/Investigator
Northwestern Medicine
Providing clinical oversight and validation for case content.
Rachel Gong
MS Machine Learning and Data Science Student
Northwestern University
Providing clinical oversight and validation for case content.
Shan Zhong
MS Machine Learning and Data Science Student
Northwestern University
Providing clinical oversight and validation for case content.
Mu He
MS Machine Learning and Data Science Student
Northwestern University
Providing clinical oversight and validation for case content.