ClinTic: Learning Clinical Dynamics through Treatment-Conditioned Neural SDEs

We propose ClinTic, a foundational model paradigm for $\text{clin}$ical trajectories built on a treatment-conditioned latent neural stochas$\text{tic}$ differential equation (SDE). Patient state evolves continuously between sparse, irregular events with treatment as the driving action; ClinTic encodes multi-modal EHR streams into the SDE latent under Markov decision process (MDP)-aligned action conditioning, and per-step decoders read out clinically meaningful targets across the rollout. An IsoFLOP sweep over seven depths and six budgets ($3\times 10^{12}$ to $10^{15}$ FLOPs) traces a clean Chinchilla-form frontier with size exponent $N^{\ast }(C)\propto C^{0.25}$, showing the paradigm is also scalable, with data more limiting than parameter count at this scale. ClinTic outperforms existing EHR foundation models and frontier LLMs across heterogeneous real-world cohorts, with the largest gain on a RECIST-PD-analogue cancer progression target (AUROC $0.827$, $+0.115$ over Opus-5). Risk representations read directly off ClinTic's dynamics separates 52-week OS where no standard biomarker reaches significance (HR $3.27$ overall, $6.35$ on Nivolumab); a further regimen-substitution probe on the same dynamics yields a stratifier of comparable discriminative power.