TEMPO: Transformers for Temporal Disease Progression from Cross-Sectional Data

arXiv:2604.23368v1 Announce Type: new Abstract: 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 extsc{Tempo}, a Transformer architecture that learns both ordinal and continuous event sequences through simulation-based supervised learning. extsc{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, extsc{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, extsc{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. extsc{Tempo} also eliminates the need to derive custom inference algorithms and enables rapid empirical comparison of generative hypotheses.

Related

• Scalable Model-Based Clustering with Sequential Monte Carlo
• Efficient Inference for Coupled Hidden Markov Models in Continuous Time and Discrete Space
• Modeling Parkinson's Disease Progression Using Longitudinal Voice Biomarkers: A Comparative Study of Statistical and Neural Mixed-Effects Models
• Online Survival Analysis: A Bandit Approach under Cox PH Model

Source: arXiv cs.LG | 2026-04-28