Disease-Progression Modelling for a Neurodegeneration Biotech
Overview
What this challenge is about.
You receive a curated longitudinal Parkinson's cohort (about 1,200 patients, 4-12 visits each, MDS-UPDRS sub-scores, cognitive assessments, demographics). Fit (1) a linear mixed-effects model with patient-specific random slopes on each progression score, (2) a state-space model that infers latent disease stage. Identify 3 candidate stratification groups (e.g., rapid vs. slow vs. cognitively-dominant) and quantify how cleanly each progression model separates them. Deliver a 4-page memo for the trial-design team with explicit caveats about cohort biases.
The Brief
What you'll do, and what you'll demonstrate.
Fit and compare two disease-progression models on a longitudinal Parkinson's cohort and propose patient-stratification groups defensible to a trial-design team.
Earning criteria — what you'll demonstrate
- Apply mixed-effects modelling to a longitudinal clinical dataset
- Fit a state-space disease-progression model and interpret latent stages
- Translate progression-model outputs into trial-design stratification proposals
- Discuss cohort-bias caveats honestly in a regulated context
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Applied AI Scientist
Bridging disease-progression modelling to trial-design stratification is the applied-AI-scientist's daily work at biotech AI teams.
This challenge sharpens
- disease-progression-modeling
- state-space-models
- model-evaluation
ML Researcher
Comparing mixed-effects and state-space models with honest cohort-bias discussion mirrors the rigour ML researchers in biotech are graded on.
This challenge sharpens
- mixed-effects-models
- state-space-models
- disease-progression-modeling
Data Scientist
Longitudinal modelling with diagnostics and stakeholder-facing memos is core senior data-scientist work in regulated industries.
This challenge sharpens
- mixed-effects-models
- ehr-modeling
- model-evaluation