# TV-based Deep 3D Self Super-Resolution for fMRI

arXiv: https://arxiv.org/abs/2410.04097 · HTML: https://mlsanity.com/paper/2410.04097/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 22 references that are in the map.
- First posted on arXiv 5 Oct 2024.

## Builds on (most cited first, 5 of 5)

- [Deep Image Prior](https://mlsanity.com/paper/1711.10925/) (2017)
- [Neural Blind Deconvolution Using Deep Priors](https://mlsanity.com/paper/1908.02197/) (2019)
- [Unsupervised Learning for Real-World Super-Resolution](https://mlsanity.com/paper/1909.09629/) (2019)
- [Self Super-Resolution for Magnetic Resonance Images using Deep Networks](https://mlsanity.com/paper/1802.09431/) (2018)
- [Super-Resolution of Brain MRI Images using Overcomplete Dictionaries and Nonlocal Similarity](https://mlsanity.com/paper/1902.04902/) (2019)

## Built on by (most cited first, 1 of 1)

- [MRI Super-Resolution with Deep Learning: A Comprehensive Survey](https://mlsanity.com/paper/2511.16854/) (2025)

## Abstract (arXiv)

While functional Magnetic Resonance Imaging (fMRI) offers valuable insights into cognitive processes, its inherent spatial limitations pose challenges for detailed analysis of the fine-grained functional architecture of the brain. More specifically, MRI scanner and sequence specifications impose a trade-off between temporal resolution, spatial resolution, signal-to-noise ratio, and scan time. Deep Learning (DL) Super-Resolution (SR) methods have emerged as a promising solution to enhance fMRI resolution, generating high-resolution (HR) images from low-resolution (LR) images typically acquired with lower scanning times. However, most existing SR approaches depend on supervised DL techniques, which require training ground truth (GT) HR data, which is often difficult to acquire and simultaneously sets a bound for how far SR can go. In this paper, we introduce a novel self-supervised DL SR model that combines a DL network with an analytical approach and Total Variation (TV) regularization. Our method eliminates the need for external GT images, achieving competitive performance compared to supervised DL techniques and preserving the functional maps.