mlsanityData to 14 Oct 2026

Integrating Mechanistic and Data-Driven Models for Neurological Disorders through Differentiable Programming

arXiv 2606.06094 ↗First posted 4 Jun 2026cs.AI cs.LG math.DScited by 0

Not on the map. Why?

Abstract (arXiv)

Advances in computational modeling, neuroimaging, and artificial intelligence are revolutionizing the modeling of neurological disorders for improved diagnostics, prognosis, and treatment planning. Mechanistic models provide valuable scientific insight into the disorders, but in practice they are often simplified with assumptions or computationally expensive and slow to solve. However, while purely data driven approaches provide speed and scalability, they require large, high quality data to train and generally suffer from interpretability and generalization issues. This perspective paper presents a structured overview of hybrid modeling strategies, which combine deep learning models with physics based solvers, and are categorized into parallel, series, and parallel-series architectures. Three main approaches that have been emphasized are residual modeling for missing or incomplete physics, Neural Ordinary Differential Equations (NODEs) for continuous time dynamics approximation, and solver in the loop that accelerates traditional solvers with neural approximations. These hybrid models integrate the governing differential equation based formulations and deep learning to characterize the evolution of neurological disorders, and promise advanced personalized neurological modeling. In addition, the study explores and proposes different hybrid configurations to improve diagnosis accuracy, predict disease progression, and inform treatment strategies across a range of neurological disorders. These capabilities outperform standalone mechanistic or purely data driven approaches, making hybrid modeling a powerful tool, especially in applications involving modeling the progression and treatment responses in neurological conditions such as brain tumors, Alzheimer's disease, and stroke.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 50 of its 132 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

50 of 132 references.

References · 50

  1. 2018Neural Ordinary Differential Equationscited by 3,779
  2. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
  3. 2015Automatic differentiation in machine learning: a surveycited by 855
  4. 2019Hamiltonian Neural Networkscited by 712
  5. 2020Neural Controlled Differential Equations for Irregular Time Seriescited by 506
  6. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 407
  7. 2020Lagrangian Neural Networkscited by 398
  8. 2019Augmented Neural ODEscited by 397
  9. 2020Universal Differential Equations for Scientific Machine Learningcited by 353
  10. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 258
  11. 2020Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solverscited by 205
  12. 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 182
38 more
  1. 2019Neural Stochastic Differential Equations: Deep Latent Gaussian Models in the Diffusion Limitcited by 176
  2. 2020Learning to Control PDEs with Differentiable Physicscited by 132
  3. 2016A Differentiable Physics Engine for Deep Learning in Roboticscited by 108
  4. 2019A Differentiable Programming System to Bridge Machine Learning and Scientific Computingcited by 83
  5. 2020NeuralSim: Augmenting Differentiable Simulators with Neural Networkscited by 82
  6. 2020Scalable Differentiable Physics for Learning and Controlcited by 73
  7. 2020On Second Order Behaviour in Augmented Neural ODEscited by 68
  8. 2020TorchMD: A deep learning framework for molecular simulationscited by 55
  9. 2024The Elements of Differentiable Programmingcited by 43
  10. 2023FluidLab: A Differentiable Environment for Benchmarking Complex Fluid Manipulationcited by 40
  11. 2024A Review of Differentiable Simulatorscited by 28
  12. 2024Differentiability in Unrolled Training of Neural Physics Simulators on Transient Dynamicscited by 27
  13. 2022Dojo: A Differentiable Physics Engine for Roboticscited by 26
  14. 2022DaXBench: Benchmarking Deformable Object Manipulation with Differentiable Physicscited by 24
  15. 2024Differentiable Programming for Differential Equations: A Reviewcited by 21
  16. 2025Comprehensive Review of Neural Differential Equations for Time Series Analysiscited by 20
  17. 2022Learning Spatio-Temporal Model of Disease Progression with NeuralODEs from Longitudinal Volumetric Datacited by 19
  18. 2023DiffMimic: Efficient Motion Mimicking with Differentiable Physicscited by 19
  19. 2022Neural Differential Equations for Learning to Program Neural Nets Through Continuous Learning Rulescited by 16
  20. 2024TorchMD-Net 2.0: Fast Neural Network Potentials for Molecular Simulationscited by 15
  21. 2021Differentiable Physics: A Position Piececited by 12
  22. 2023Predictive Digital Twin for Optimizing Patient-Specific Radiotherapy Regimens under Uncertainty in High-Grade Gliomascited by 12
  23. 2024BrainODE: Dynamic Brain Signal Analysis via Graph-Aided Neural Ordinary Differential Equationscited by 12
  24. 2018The multiphysics of prion-like diseases: progression and atrophycited by 6
  25. 2024Universal Differential Equations as a Common Modeling Language for Neurosciencecited by 5
  26. 2024Non-Negative Universal Differential Equations With Applications in Systems Biologycited by 5
  27. 2019Differentiable programming and its applications to dynamical systemscited by 3
  28. 2021Distinguishing Healthy Ageing from Dementia: a Biomechanical Simulation of Brain Atrophy using Deep Networkscited by 3
  29. 2022AI-enhanced iterative solvers for accelerating the solution of large scale parametrized systemscited by 3
  30. 2024Stabilizing Backpropagation Through Time to Learn Complex Physicscited by 3
  31. 2025PRDP: Progressively Refined Differentiable Physicscited by 3
  32. 2025Mechanistic Learning with Guided Diffusion Models to Predict Spatio-Temporal Brain Tumor Growthcited by 3
  33. 2024Accelerated Patient-Specific Calibration via Differentiable Hemodynamics Simulationscited by 2
  34. 2020Temporal Dynamic Model for Resting State fMRI Data: A Neural Ordinary Differential Equation approachcited by 1
  35. 2021Physics-informed neural networks for improving cerebral hemodynamics predictionscited by 1
  36. 2024A Step-by-step Introduction to the Implementation of Automatic Differentiationcited by 1
  37. 2025Exploring Neural Ordinary Differential Equations as Interpretable Healthcare classifierscited by 1
  38. 2025Missing Physics Discovery through Fully Differentiable Finite Element-Based Machine Learningcited by 1

Built on by

Not cited yet.