Smart Clinical Assistance: Decision Support
LENS (ScaDS.AI-Seedfunding)
Learning Engine for Novel Survival-clustering (ScaDS.AI seed funding)
Learning Engine for Novel Survival-clustering: a model-agnostic method that finds prognostically distinct patient groups – pioneered on multiple myeloma data and CT images of non-small cell lung cancer.
Risk stratification is vital for clinical decisions, but existing methods often don’t convert advanced survival analysis into actionable criteria. LENS aims to introduce a model-agnostic method that can train any neural network on any data modality to find prognostically distinct patient groups by directly optimizing survival heterogeneity across clusters.
The approach is pioneered on real-world cancer datasets from multiple myeloma and CT images from non-small cell lung cancer. External and institutional validations support the method’s reliability. The pan-cancer, modality-independent framework facilitates discovery of prognostic signatures and produces interpretable outputs aimed at supporting personalized treatment and clinical decision-making.
More projects of the team
-
VOLTA Validation Of Model-Based Therapy Decision Support in Hematology
Validation of model-based therapy decision support in hematology: KAIT, the platform for multiple myeloma developed with the hematology department at Leipzig University Hospital since 2020, is being expanded, tested and optimized for routine practice.
-
GenoMed4All
EU-wide network for exchanging clinical and multi-omics data via federated learning for hematological diseases; ICCAS contributes data integration and standardization based on HL7 FHIR.
Back to the team Smart Clinical Assistance: Decision Support