Examples

Examples of current research work in the group "AI & Data Science"

Alzheimer early
detection

Retrospectively validated AI model allows early detection of Alzheimer's 6 years in advance
 

Molecular subtypes

 

Innovative AI approach allows identification of prognostically relevant molecular cancer subtypes based on multi-omics data

 

Drug target

 

AI approach allows more efficient identification of new drug target structures
 

Synthetic data

 

Innovative AI approach allows the generation of realistic synthetic patient trajectories
 

Mechanisms based patient stratification

AI allows for stratification of Alzheimer and Parkinson patients via molecular disease mechanisms - a step towards Precision medicine in neurology

 

Progression Subtypes

 

New AI approach allows for clustering of multivariate longitudinal patient trajectories and detection of progression subtypes in Alzheimer's and Parkinson's Disease

 

 

Explainable AI

 

An explainable AI approach allows for predicting comorbidity risks of individual epileptic patients using large scale clinical routine data. Within the Fraunhofer Center for Machine Learning we developed a demonstrator to showcase this approach.

Predicting drug response

 

Towards realizing the vision of precision medicine: AI allows for predicting response to anti-epileptic drug