Challenge Overview: It is often very difficult for patients to find clinical trials they can participate in, and this inability to quickly identify and match patients to clinical studies often results in delays that can jeopardize patient lives. As the Life Sciences industry is investing more heavily in rare diseases, it’s becoming critically important to the success of these research programs to find those rare patients who are willing and able to participate in these studies. Adding to the complexity, patients may be ineligible to participate due to certain elements of their medical history. In this challenge, you will be showing how you can extract inclusion/exclusion (I/E) criteria from publicly available clinical trials registries and then develop a chatbot patients can use to find studies and then complete pre-screening questionnaires to determine if they are a match.
Approach: 1.Pull listings of Non-Small Cell Lung Cancer (NSCLC) studies from clinicaltrials.gov with a status of recruiting. 2.Use this link to pull corresponding studies: Search of: Recruiting Studies | NSCLC | Phase 3 | Industry - List Results - ClinicalTrials.gov 3.Pull inclusion/exclusion criteria for each study. You will need to store these criteria in a database. 4.Develop an AI-generated prescreening chat interface that uses the I/E criteria to help NSCLS patients find trials that are enrolling and that may be a match for them. 5.Note: The I/E criteria can become very tricky (it may focus on stage of disease or biomarker presence, among other things, that would be difficult to develop during a hackathon). For that reason, focus on a small subset of the criterias. As an example, you can consume only I/E criteria related to age, location of the patient, medications and history of other primary metastatic conditions.
Log in or sign up for Devpost to join the conversation.