Work Package 7: Data Analytics and Machine Learning

Machine Learning for Subject & Cohort Analysis 

The objectives of Work Package 7 are to develop 1) segmentation models to identify cohorts of patients suitable for clinical trials, 2) data-driven prediction models of ALS progression, and 3) data exploration and analysis methods to generate clinical insight into ALS. This WP will leverage the Precision ALS platform developed in WP1 to apply state-of-the-art data analysis techniques and address key clinical challenges in ALS research. 

A central aim is to facilitate the identification of subcohorts of patients that cannot be currently identified using clinical evaluation alone. Cohortisation of patients can be achieved through multi-modal analysis of patient types, based on data collected in Work Package 2. Disease progression analysis will form another central objective. This will involve the employment of the variety of modelling data available, as well as scan and imaging data to construct a predictive model of ALS progression. The long-term goal of this work package is novel insight generation to complement future research. Insights can be generated by analysing markers of interest and generating models to identify informative patterns in the data of ALS patients.

The Precision ALS platform will store a variety of data, taken at multiple timepoints. Therefore, this work package aims to enhance the accuracy of data analysis in both the context of varying datatypes, and the context of data that changes over time.

Leaders: Prof. John Kelleher