Our Goal is to help cancer patients find clinical trials that are the best fit for their diagnosis in a more user friendly way - expanding access, particularly to the underserved. In response, we created an AI enhanced conversational AI platform (think chatbot!) that simplifies cancer clinical trial searches. We transform complex queries into natural dialogues using open source Llama3-8b-8192 AI model, tailored for medical contexts. Our platform combines advanced medical language understanding, dynamic query construction, and real-time integration with Clinical Trials (ct.gov) API v2 to deliver personalized, location-aware recommendations. Built on a robust stack with Next.js, TypeScript, and Tailwind CSS, it ensures optimal performance through geo-spatial relevance, interactive guidance, and secure, efficient data handling. We will expand this approach to other domains including labor, the environment, and agriculture.
To-Be
Since this is still in MVP phase, we do not have any specific out comes. However, based on our initial user testing, we have received great feedback, on its ease of use and utility - much improved over standard faceted searches.
Stakeholders
We are currently in discussions with the National Cancer Institute and our client at hhs.gov.
Business Process Model
Automating Manual Processes – Replacing repetitive manual workflows with automated systems to improve efficiency and reduce errors.;Improving Decision-Making with Actionable Insights – Enabling more informed decisions by providing clear, actionable, and data-driven insights to stakeholders.;Allows federal agencies to improve customer/patient experience for constituents, via the use of AI - expanding the reach and effectiveness of searches/data that are otherwise difficult to mine..