AIPD-DC9-PETA

• Host Institution: Petanux
  
• PhD Enrolment: University of Luxembourg (UL)
  
• Start Date: October 2025
  
• Duration: 36 months
• Official PhD Supervisor: Jochen Klucken

Research Objectives

UL has developed a new clinical PD staging system, based on the requirements of healthcare experts, to support personalized clinical decision-making and the selection of patient-tailored treatment goals for everyday healthcare workups and procedures. The goal of this research project is to develop an AI-based prediction of healthcare-related PD stages. Such prediction tools are of increasing relevance in a digital-supported healthcare system, to trigger the adaptation of current treatment plans and care pathways, as well as to provide early preventative strategies to high-risk patients. Additionally, prediction of stages will directly impact healthcare systems by improving resource allocation and the need for care procedures reimbursement. We will employ multi-modal data from LuxPARK (patient reported outcome measures, clinical scores, gait biomarkers) to develop classical (Random Forest, XGBoost) as well as neural network (e.g. LSTM) based AI/ML classifiers. XAI techniques such as SHAP will be employed to provide a human understandable model explanation. This project complements the work of DCs 2 and 12, which focus on disease progression from the angle of clinical research data and digital gait biomarkers, respectively. DC9 in contrast, focuses more on the current healthcare needs. Accordingly, DC9 will collaborate with DCs 2 and 12.

Expected Results

• Novel AI/ML algorithms predicting healthcare-related PD stages
• Better understanding of disease trajectories
   

Planned Secondment(s)

• Host: University of Luxembourg
  
  • Duration: 18 months
    
  • Purpose: Learning about PD disease management and LuxPARK data

This project is part of the "Digital Health" work package.