Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training

Aurora Pia Ghiardelli; Guangzhi Tang; Tao Sun

doi:10.1109/ISBI61048.2026.11516057

Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training

Aurora Pia Ghiardelli
, Guangzhi Tang
, Tao Sun^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

Abstract

We propose a reliable and energy-efficient framework for 3D brain tumor segmentation using spiking neural networks (SNNs). A multi-view ensemble of sagittal, coronal, and axial SNN models provides voxel-wise uncertainty estimation and enhances segmentation robustness. To address the high computational cost in training SNN models for semantic image segmentation, we employ Forward Propagation Through Time (FPTT), which maintains temporal learning efficiency with significantly reduced computational cost. Experiments on the Multimodal Brain Tumor Segmentation Challenges (BraTS 2017 and BraTS 2023) demonstrate competitive accuracy, well-calibrated uncertainty, and an 87 % reduction in FLOPs, underscoring the potential of SNNs for reliable, low-power medical IoT and Point-of-Care systems.

Original language	English
Title of host publication	ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging
Publisher	IEEE Computer Society
Volume	2026-April
ISBN (Electronic)	9798331577636
DOIs	https://doi.org/10.1109/ISBI61048.2026.11516057
Publication status	Published - 2026
Event	23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026 - London, United Kingdom Duration: 8 Apr 2026 → 11 Apr 2026 https://biomedicalimaging.org/2026/

Publication series

Series	Proceedings - International Symposium on Biomedical Imaging
Volume	2026-April
ISSN	1945-7928

Conference

Conference	23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026
Abbreviated title	ISBI 2026
Country/Territory	United Kingdom
City	London
Period	8/04/26 → 11/04/26
Internet address	https://biomedicalimaging.org/2026/

Keywords

Brain Tumor Segmentation
Forward Propagation Through Time (FPTT)
Spiking Neural Networks (SNNs)
Uncertainty Estimation

Access to Document

10.1109/ISBI61048.2026.11516057

Cite this

@inproceedings{5a63d99741d7436dafe74ece84fe42a4,

title = "Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training",

abstract = "We propose a reliable and energy-efficient framework for 3D brain tumor segmentation using spiking neural networks (SNNs). A multi-view ensemble of sagittal, coronal, and axial SNN models provides voxel-wise uncertainty estimation and enhances segmentation robustness. To address the high computational cost in training SNN models for semantic image segmentation, we employ Forward Propagation Through Time (FPTT), which maintains temporal learning efficiency with significantly reduced computational cost. Experiments on the Multimodal Brain Tumor Segmentation Challenges (BraTS 2017 and BraTS 2023) demonstrate competitive accuracy, well-calibrated uncertainty, and an 87 \% reduction in FLOPs, underscoring the potential of SNNs for reliable, low-power medical IoT and Point-of-Care systems.",

keywords = "Brain Tumor Segmentation, Forward Propagation Through Time (FPTT), Spiking Neural Networks (SNNs), Uncertainty Estimation",

author = "Ghiardelli, \{Aurora Pia\} and Guangzhi Tang and Tao Sun",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026, ISBI 2026 ; Conference date: 08-04-2026 Through 11-04-2026",

year = "2026",

doi = "10.1109/ISBI61048.2026.11516057",

language = "English",

volume = "2026-April",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

booktitle = "ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging",

address = "United States",

url = "https://biomedicalimaging.org/2026/",

}

Ghiardelli, AP, Tang, G & Sun, T 2026, Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training. in ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging. vol. 2026-April, IEEE Computer Society, Proceedings - International Symposium on Biomedical Imaging, vol. 2026-April, 23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026, London, United Kingdom, 8/04/26. https://doi.org/10.1109/ISBI61048.2026.11516057

Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training. / Ghiardelli, Aurora Pia; Tang, Guangzhi ; Sun, Tao.
ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging. Vol. 2026-April IEEE Computer Society, 2026. (Proceedings - International Symposium on Biomedical Imaging, Vol. 2026-April).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

TY - GEN

T1 - Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training

AU - Ghiardelli, Aurora Pia

AU - Tang, Guangzhi

AU - Sun, Tao

PY - 2026

Y1 - 2026

N2 - We propose a reliable and energy-efficient framework for 3D brain tumor segmentation using spiking neural networks (SNNs). A multi-view ensemble of sagittal, coronal, and axial SNN models provides voxel-wise uncertainty estimation and enhances segmentation robustness. To address the high computational cost in training SNN models for semantic image segmentation, we employ Forward Propagation Through Time (FPTT), which maintains temporal learning efficiency with significantly reduced computational cost. Experiments on the Multimodal Brain Tumor Segmentation Challenges (BraTS 2017 and BraTS 2023) demonstrate competitive accuracy, well-calibrated uncertainty, and an 87 % reduction in FLOPs, underscoring the potential of SNNs for reliable, low-power medical IoT and Point-of-Care systems.

AB - We propose a reliable and energy-efficient framework for 3D brain tumor segmentation using spiking neural networks (SNNs). A multi-view ensemble of sagittal, coronal, and axial SNN models provides voxel-wise uncertainty estimation and enhances segmentation robustness. To address the high computational cost in training SNN models for semantic image segmentation, we employ Forward Propagation Through Time (FPTT), which maintains temporal learning efficiency with significantly reduced computational cost. Experiments on the Multimodal Brain Tumor Segmentation Challenges (BraTS 2017 and BraTS 2023) demonstrate competitive accuracy, well-calibrated uncertainty, and an 87 % reduction in FLOPs, underscoring the potential of SNNs for reliable, low-power medical IoT and Point-of-Care systems.

KW - Brain Tumor Segmentation

KW - Forward Propagation Through Time (FPTT)

KW - Spiking Neural Networks (SNNs)

KW - Uncertainty Estimation

U2 - 10.1109/ISBI61048.2026.11516057

DO - 10.1109/ISBI61048.2026.11516057

M3 - Conference article in proceeding

VL - 2026-April

T3 - Proceedings - International Symposium on Biomedical Imaging

BT - ISBI 2026 - 23rd IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

T2 - 23rd IEEE International Symposium on Biomedical Imaging, ISBI 2026

Y2 - 8 April 2026 through 11 April 2026

ER -

Reliable Brain Tumor Segmentation Based on Spiking Neural Networks with Efficient Training

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this