Version 8.1 by mhashemi on 2024/11/27 18:03
Hide last authors
mhashemi 8.1 1 (% style="text-align: justify;" %)
2 {{{The integration of an individual’s brain imaging data in VBMs has improved patient-specific predictivity, although Bayesian
mhashemi 7.1 3 estimation of spatially distributed parameters remains challenging even with state-
4 of-the-art Monte Carlo sampling. VBMs imply latent nonlinear state space models
5 driven by noise and network input, necessitating advanced probabilistic machine
6 learning techniques for widely applicable Bayesian estimation. Here we present
7 Simulation-based Inference on Virtual Brain Models (SBI-VBMs), and demonstrate
8 that training deep neural networks on both spatio-temporal and functional fea-
9 tures allows for accurate estimation of generative parameters in brain disorders.
10 The systematic use of brain stimulation provides an effective remedy for the non-
11 identifiability issue in estimating the degradation limited to smaller subset of con-
12 nections. By prioritizing model structure over data, we show that the hierarchical
13 structure in SBI-VBMs renders the inference more effective, precise and biologically
14 plausible. This approach could broadly advance precision medicine by enabling fast
15 and reliable prediction of patient-specific brain disorders.
mhashemi 1.1 16
mhashemi 7.1 17 Code:
18 }}}
mhashemi 1.1 19
20
mhashemi 8.1 21 (% style="text-align: justify;" %)
mhashemi 7.1 22 Ref: [[https:~~/~~/iopscience.iop.org/article/10.1088/2632-2153/ad6230>>https://iopscience.iop.org/article/10.1088/2632-2153/ad6230]]