Inferring appropriate eligibility criteria in clinical trial protocols without labeled data

Abstract

We consider the user task of designing clinical trial protocols and propose a method that outputs the most appropriate eligibility criteria from a potentially huge set of candidates. Each document d in our collection D is a clinical trial protocol which itself contains a set of eligibility criteria. Given a small set of sample documents D', |D'|<<|D|, a user has initially identified as relevant e.g., via a user query interface, our scoring method automatically suggests eligibility criteria from D by ranking them according to how appropriate they are to the clinical trial protocol currently being designed. We view a document as a mixture of latent topics and our method exploits this by applying a three-step procedure. First, we infer the latent topics in the sample documents using Latent Dirichlet Allocation (LDA) [3]. Next, we use logistic regression models to compute the probability that a given candidate criterion belongs to a particular topic. Lastly, we score each criterion by computing its expected value, the probability-weighted sum of the topic proportions inferred from the set of sample documents. Intuitively, the greater the probability that a candidate criterion belongs to the topics that are dominant in the samples, the higher its expected value or score. Results from our experiments indicate that our proposed method is 8 and 9 times better (resp., for inclusion and exclusion criteria) than randomly choosing from a set of candidates obtained from relevant documents. In user simulation experiments, we were able to automatically construct eligibility criteria that are on the average 75% and 70% (resp., for inclusion and exclusion criteria) similar to the correct eligibility criteria.