TEMPO: Transformers for Temporal Disease Progression from Cross-Sectional Data
For researchers modeling disease progression from cross-sectional data, TEMPO provides a more accurate and flexible alternative to Event-Based Models, eliminating the need for custom inference algorithms.
TEMPO is a Transformer-based model that learns both ordinal and continuous event sequences from cross-sectional data for disease progression modeling. It reduces normalized Kendall's Tau distance by 52.89% and staging MAE by 25.33% over state-of-the-art SA-EBM on synthetic benchmarks, and recovers a biologically plausible Alzheimer's progression on ADNI data.
Event-Based Models (EBMs) infer biomarker progression from cross-sectional data but typically only as ordinal sequences and rely on rigid model assumptions. We propose \textsc{Tempo}, a Transformer architecture that learns both ordinal and continuous event sequences through simulation-based supervised learning. \textsc{Tempo} uses two Transformer modules: one treats biomarkers as tokens to infer event sequencing; the other treats patients as tokens, representing each by their per-biomarker abnormality profile, to infer patients' disease stages. On synthetic benchmarks, \textsc{Tempo} reduces normalized Kendall's Tau distance by 52.89\% and staging MAE by 25.33\% compared to state-of-the-art SA-EBM, with larger reductions in high-dimensional settings (58.88\% and 61.10\%). Applied to ADNI, \textsc{Tempo} recovers a biologically plausible Alzheimer's progression: early medial temporal atrophy, followed by amyloid accumulation and cognitive decline, and late-stage tau pathology with terminal acceleration of global neurodegeneration -- broadly consistent with established disease models. \textsc{Tempo} also eliminates the need to derive custom inference algorithms and enables rapid empirical comparison of generative hypotheses.