Related Experiment Video
Updated: May 12, 2026
08:41
Analyzing Dendritic Morphology in Columns and Layers
Published on: March 23, 2017
9.4K
NIEND: neuronal image enhancement through noise disentanglement
Zuo-Han Zhao1, Lijuan Liu1, Yufeng Liu1
1SEU-ALLEN Joint Center, Institute for Brain and Intelligence, Southeast University, Nanjing, Jiangsu 210096, China.
Bioinformatics (Oxford, England)
|March 26, 2024
Summary
We developed Neuronal Image Enhancement through Noise Disentanglement (NIEND) to improve noisy neuronal images. NIEND significantly enhances image quality and neuron reconstruction accuracy for automated analysis.
Area of Science:
- Neuroscience
- Computational Biology
- Image Processing
Background:
- Automated digital neuronal reconstruction is hindered by noisy light microscopic images.
- Existing methods struggle to balance robustness and computational efficiency for image quality improvement.
Purpose of the Study:
- To introduce Neuronal Image Enhancement through Noise Disentanglement (NIEND), a novel pipeline for enhancing neuronal images.
- To improve the quality of neuronal images for more accurate automated reconstruction.
Main Methods:
- Developed the NIEND image enhancement pipeline using Python 3.10.
- Benchmarked NIEND on 863 mouse neuronal images with gold standards.
- Implemented NIEND with Vaa3D compatibility and made it publicly available on GitHub.
Main Results:
- NIEND achieved 40-fold improvement in signal-background contrast and 10-fold in background uniformity.
- Average F1 score for neuron reconstruction improved to 0.88 from 0.78, outperforming other methods.
- NIEND processes large images rapidly (1.6s for 256^3 images) with significant data compression (1% by LZMA).
Conclusions:
- NIEND substantially improves neuronal image quality and automated reconstruction accuracy.
- The method offers a robust and computationally efficient solution for large-scale neuronal analysis.
- NIEND has the potential to advance automated neuron morphology reconstruction for petascale datasets.
