This project is part of the AMP SCZ program, an initiative for early detection of risk for schizophrenia(https://www.ampscz.org).
A key goal in AMPSCZ is to predict which patients that present initially mild or sub-threshold symptoms will eventually develop psychosis. Most predictive models are based on data acquired on their first medical visit (the baseline visit). An important question is how much is gained by following patients over time (longitudinal data). In this project we will implement predictive models that make use of this longitudinal information for psychosis prediction. We will focus on implementing a type of models called "joint models", which incorporate time-varying predictors into well known survival analyses.
Objective
Objective A. Implement a Python-based version of longitudinal models adapted for common best practices in machine learning (separate train/test, scikit-learn compatible methods).
Objective B Quantify the advantage of longitudinal models vs baseline predictors in a legacy dataset.
Approach and Plan
Write a python wrapper, using rpy2, for the R library JM that implements longitudinal analysis.
Use synthetic and legacy datasets to test the predictions.
Use python libraries such as lifelines or scikit-survival to implement survival analysis with baseline predictions only.
Implement permutation tests in time to asses the significance of prediction improvements due to longitudinal change.
Category
Quantification and Computation
Key Investigators
Project Description
This project is part of the AMP SCZ program, an initiative for early detection of risk for schizophrenia(https://www.ampscz.org).
A key goal in AMPSCZ is to predict which patients that present initially mild or sub-threshold symptoms will eventually develop psychosis. Most predictive models are based on data acquired on their first medical visit (the baseline visit). An important question is how much is gained by following patients over time (longitudinal data). In this project we will implement predictive models that make use of this longitudinal information for psychosis prediction. We will focus on implementing a type of models called "joint models", which incorporate time-varying predictors into well known survival analyses.
Objective
Approach and Plan
Progress and Next Steps
Illustrations
No response
Background and References