Abstract
Deeplearning-basedspatio-temporaldiseaseprogressionmodels commonly overlook the incubation period of progressive diseases, limiting the use of those models in early interventions, which are vitalfornoteasilyreversiblediseasessuchasAlzheimer’s.Thisisbecause, the existing deep learning based longitudinal disease-progression models are almost always run forward: from an observed baseline they predict future decline. In many clinical settings, however, imaging begins onlyafterpathologyissuspectedoralreadyvisible,theearlier,healthier patient-specific reference was never acquired. To address this, we proposetostudyreverse disease progression prediction:givenlaterdiseased anatomy,reconstructtheunobservedhealthieranatomythatprecededit. Weuseatwo-stagemodelinwhichafrozen3Dvector-quantisedautoencoder defines a compact discrete latent space, while a Neural Ordinary Differential Equation (ODE) learns continuous-time dynamics in that space. A recurrent encoder reads late observations in reverse temporal order, initialises the latent state, and the ODE is integrated backwards acrossthetrajectory.OnacontrolledMorpho-MNISTbenchmarkwitha sinusoidalperturbation,ourmodelsuccessfullyrecoveredtheunseenpreviousstatesfromlaterobservationsofthenon-monotonictrajectory.On longitudinal brain MRIs from Alzheimer’s Disease Neuroimaging Initiative,atthetasktorecoverthepreviousunseentrajectorytowardshealthy statesofthepatientsfromobservedlaterdiseasedstates,ourmodeloutperforms the baselines that uses copy-nearest and mean-observed, with positivedisease-reversalscoresineverydiagnosticstratum.Wehopethat our work can provideinsights and tools towards discovering the incubation periods from single-shot scans, and developing early interventions of diseases based on imaging.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/SASHIMI_031.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=hKUOEhKgMH
BibTex
@InProceedings{SheUlu_Reverse_MICCAISAT2026,
author = { Shernazarov, Ulugbek AND Xu, Moucheng AND Ramatov, Inomjon},
title = { { Reverse Spatio-Temporal Disease Progression Modelling } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17258},
month = {pending},
page = {pending}
}