# Neural Pre-Processing: A Learning Framework for End-to-end Brain MRI Pre-processing

Source: [Neural Pre-Processing: A Learning Framework for End-to-end Brain MRI Pre-processing](<https://www.sota2.com/research/paper/68fa6ecd1e209131cea4239e>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T03:47:50.366Z

Authors: Xinzi He, Alan Wang, Mert R. Sabuncu

Published: 2023-03-21T19:10:21.000Z

## Abstract

Head MRI pre-processing involves converting raw images to an intensity-normalized, skull-stripped brain in a standard coordinate space. In this paper, we propose an end-to-end weakly supervised learning approach, called Neural Pre-processing (NPP), for solving all three sub-tasks simultaneously via a neural network, trained on a large dataset without individual sub-task supervision. Because the overall objective is highly under-constrained, we explicitly disentangle geometric-preserving intensity mapping (skull-stripping and intensity normalization) and spatial transformation (spatial normalization). Quantitative results show that our model outperforms state-of-the-art methods which tackle only a single sub-task. Our ablation experiments demonstrate the importance of the architecture design we chose for NPP. Furthermore, NPP affords the user the flexibility to control each of these tasks at inference time. The code and model are freely-available at \\url{https://github.com/Novestars/Neural-Pre-processing}.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Skull-stripping on Brain MRI 256x256x256](<https://www.sota2.com/research/sota/skull-stripping-on-brain-mri-256x256x256.md>): Skull-stripping; Brain MRI 256x256x256; Average Runtime (s): 2.94.

- [Intensity Normalization on Brain MRI 256x256x256](<https://www.sota2.com/research/sota/intensity-normalization-on-brain-mri-256x256x256.md>): Intensity Normalization; Brain MRI 256x256x256; Avg Runtime (sec): 2.94.

- [Intensity Normalization on HCP](<https://www.sota2.com/research/sota/intensity-normalization-on-hcp.md>): Intensity Normalization; HCP; Reconstruction SSIM: 99.1.

- [Spatial Normalization on Brain MRI 256x256x256](<https://www.sota2.com/research/sota/spatial-normalization-on-brain-mri-256x256x256.md>): Spatial Normalization; Brain MRI 256x256x256; Average Runtime (sec): 2.94.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2303.12148>)

[PDF](<https://arxiv.org/pdf/2303.12148.pdf>)

[Implementation](<https://github.com/Novestars/Neural-Pre-processing>)