The task for matching patients requires searching for appropriate clinical trials and researchers recruiting for their studies require analysis of patient and trial criteria. MARUL is designed to consider the information available from trials and constructs personalized surveys for patients to match them with the best trials applicable to them.
Technology Overview
This novel system employs natural language processing and machine learning techniques to address this challenging task. System design for MARUL can be described as five components: 
- Information extraction from trial data 
- Clustering inclusion and exclusion criteria 
- Classification of criteria types 
- Building personalized surveys 
- Estimation of matching score
- Transformation from textual data to structured databases
- Generating short surveys that can capture the most essential information about patients conditions 
- Assignment of patients to the appropriate trials more efficiently
- Data mining over large clinical texts
- Combining biological data with the user surveys in a way to complement each other 
- Building personalized surveys to match with clinical trials
- License
- Partnering
- Research collaboration
Patent Information:
For Information, Contact:
Mark Saulich
Associate Director of Commercialization
Northeastern University
Albert-Laszlo Barabasi
Computational Modeling
Data Mining
Machine Learning