3D Neuron Reconstruction in Tangled Neuronal Image With Deep Networks

Li, Qiufu; Shen, Linlin

doi:10.1109/tmi.2019.2926568

Cited by 57 publications

(34 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, in this method, a single neuron is segmented from the overlapping neuron structure based on a weakly label, producing improved results without intersecting points in tracing [20].…”

Section: Research Of Neuron Structurementioning

confidence: 99%

Neuron Segmentation using Incomplete and Noisy Labels via Adaptive Learning with Structure Priors

Park

Lee

Kim

et al. 2021

2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)

View full text Add to dashboard Cite

Recent advances in machine learning have demonstrated significant success in biomedical image segmentation. Most existing high-quality segmentation algorithms rely on supervised learning with full training labels. However, segmentation is more susceptible to label quality; notably, generating accurate labels in biomedical data is a labor-and time-intensive task. Especially, structure neuronal images are hard to obtain full annotation because of the entangled shape of each structure. In this thesis, a neuron structure semantic segmentation algorithm is proposed on a noise label. I assume that the label has noise and propose two new novel loss functions.Adaptive loss is applied to noise pixels in different labels with prediction in partially annotated labels. These fluorescence images may have confidence that can leverage prior knowledge when each pixel has intensity. Reconstruction loss is suggested that can be regularized of neuronal cell structures to reduce false segmentation near noisy labels. Additionally, This study is aimed to verify that our method preserves the connectivity of linear structure through a novel evaluation matrix.

show abstract

Section: Research Of Neuron Structurementioning

confidence: 99%

Neuron Segmentation using Incomplete and Noisy Labels via Adaptive Learning with Structure Priors

Park

Lee

Kim

et al. 2021

2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)

View full text Add to dashboard Cite

show abstract

“…A natural way to improve the performance of existing neuron tracing algorithms is to extract the complete neuron from the noisy image first and then automatically reconstruct it. Deep learning, which achieves excellent results in many fields, has also been applied into 3D neuron segmentation ( Huang et al , 2020 ; Jiang et al , 2021 ; Klinghoffer et al , 2020 ; Li et al , 2017 ; Li and Shen, 2020 ). As nerve fibers can spread in a very large brain region, a large neuronal image has to be processed for segmenting the complete neuron, which leads to great computational cost for the deep networks-based 3D neuron segmentation.…”

Section: Introductionmentioning

confidence: 99%

Neuron segmentation using 3D wavelet integrated encoder–decoder network

Shen

2021

Bioinformatics

Self Cite

View full text Add to dashboard Cite

Motivation 3D neuron segmentation is a key step for the neuron digital reconstruction, which is essential for exploring brain circuits and understanding brain functions. However, the fine line-shaped nerve fibers of neuron could spread in a large region, which brings great computational cost to the neuron segmentation. Meanwhile, the strong noises and disconnected nerve fibers bring great challenges to the task. Results In this paper, we propose a 3D wavelet and deep learning based 3D neuron segmentation method. The neuronal image is first partitioned into neuronal cubes to simplify the segmentation task. Then, we design 3D WaveUNet, the first 3D wavelet integrated encoder-decoder network, to segment the nerve fibers in the cubes; the wavelets could assist the deep networks in suppressing data noises and connecting the broken fibers. We also produce a Neuronal Cube Dataset (NeuCuDa) using the biggest available annotated neuronal image dataset, BigNeuron, to train 3D WaveUNet. Finally, the nerve fibers segmented in cubes are assembled to generate the complete neuron, which is digitally reconstructed using an available automatic tracing algorithm. The experimental results show that our neuron segmentation method could completely extract the target neuron in noisy neuronal images. The integrated 3D wavelets can efficiently improve the performance of 3D neuron segmentation and reconstruction. Availability The data and codes for this work are available at https://github.com/LiQiufu/3D-WaveUNet. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

“…Machine-learning based methods are proposed to improve the foreground estimation from uneven neuronal images ( Frasconi et al, 2014 ; Chen et al, 2015 ; Li R. et al, 2017 ; Mazzamuto et al, 2018 ; Li and Shen, 2020 ). Support vector machine (SVM) based methods use careful designed hand-crafted features to identify foreground signals whose features are consistent with the training set ( Chen et al, 2015 ; Li S. et al, 2019 ), and improve the accuracy in inhomogeneous neurite tracing.…”

Section: Introductionmentioning

confidence: 99%

“…These methods are also computational complex and time-consuming, and they may need to construct corresponding training set for every sub-stack in a large-scale dataset. Deep-learning based methods ( Frasconi et al, 2014 ; Li R. et al, 2017 ; Mazzamuto et al, 2018 ; Li and Shen, 2020 ; Huang et al, 2021 ) have been employed for cell identification and neuron reconstruction. These methods use deep convolutional network to extract deeper and more abundant features of neuronal images, and some also take advantage of traditional methods, such as mean-shift, Isomap algorithm ( Frasconi et al, 2014 ), probabilistic blob detection ( Mazzamuto et al, 2018 ), and content-aware adaptive voxel scooping ( Huang et al, 2021 ), together to boost the accuracy in the foreground estimation from various neuronal images and even large-scale datasets significantly.…”

Section: Introductionmentioning

confidence: 99%

Foreground Estimation in Neuronal Images With a Sparse-Smooth Model for Robust Quantification

et al. 2021

View full text Add to dashboard Cite

3D volume imaging has been regarded as a basic tool to explore the organization and function of the neuronal system. Foreground estimation from neuronal image is essential in the quantification and analysis of neuronal image such as soma counting, neurite tracing and neuron reconstruction. However, the complexity of neuronal structure itself and differences in the imaging procedure, including different optical systems and biological labeling methods, result in various and complex neuronal images, which greatly challenge foreground estimation from neuronal image. In this study, we propose a robust sparse-smooth model (RSSM) to separate the foreground and the background of neuronal image. The model combines the different smoothness levels of the foreground and the background, and the sparsity of the foreground. These prior constraints together contribute to the robustness of foreground estimation from a variety of neuronal images. We demonstrate the proposed RSSM method could promote some best available tools to trace neurites or locate somas from neuronal images with their default parameters, and the quantified results are similar or superior to the results that generated from the original images. The proposed method is proved to be robust in the foreground estimation from different neuronal images, and helps to improve the usability of current quantitative tools on various neuronal images with several applications.

show abstract

3D Neuron Reconstruction in Tangled Neuronal Image With Deep Networks

Cited by 57 publications

References 41 publications

Neuron Segmentation using Incomplete and Noisy Labels via Adaptive Learning with Structure Priors

Neuron Segmentation using Incomplete and Noisy Labels via Adaptive Learning with Structure Priors

Neuron segmentation using 3D wavelet integrated encoder–decoder network

Foreground Estimation in Neuronal Images With a Sparse-Smooth Model for Robust Quantification

Contact Info

Product

Resources

About