Identifying Critical Phases for Disease Onset with Sparse Haematological Biomarkers
This addresses the need for interpretable disease prediction models in clinical settings using sparse data, representing an incremental improvement over traditional imputation methods.
The paper tackled the problem of predicting disease onset from irregularly sampled blood test data by proposing Graph Neural Additive Networks (GNAN) to model biomarker trajectories without imputation, achieving interpretable predictions and preserving temporal structure.
Routinely collected clinical blood tests are an emerging molecular data source for large-scale biomedical research but inherently feature irregular sampling and informative observation. Traditional approaches rely on imputation, which can distort learning signals and bias predictions while lacking biological interpretability. We propose a novel methodology using Graph Neural Additive Networks (GNAN) to model biomarker trajectories as time-weighted directed graphs, where nodes represent sampling events and edges encode the time delta between events. GNAN's additive structure enables the explicit decomposition of feature and temporal contributions, allowing the detection of critical disease-associated time points. Unlike conventional imputation-based approaches, our method preserves the temporal structure of sparse data without introducing artificial biases and provides inherently interpretable predictions by decomposing contributions from each biomarker and time interval. This makes our model clinically applicable, as well as allowing it to discover biologically meaningful disease signatures.