About this Research Topic
Our brain is composed of tens of billions of neurons, connected with each other forming neural circuits. Neurites are often so thin that the diameter is measurable on the scale of tens of nanometers, while axons, on the other hand, can be so long that they could project to another hemisphere. This is known as the “big and small” challenge.
Benefiting from recent breakthroughs in imaging techniques, big samples could be imaged with high resolution. With light microscopy, a whole mouse or rhesus monkey brain could be imaged at the level of single neurite resolution. Electron microscopy allows on the other hand to image a cubic millimeter of brain tissue at an unparalleled nanometer resolution, consenting to resolve synapses and trace nanometer-thin neuronal processes. Other imaging methods, such as Synchrotron-Based X-ray Imaging, are improving and have already produced impressive images.
New tissue clearing systems enable efficient imaging of high-resolution individual nerve axons in the brain, spinal cord, and peripheral organs. A variety of large-scale image volumes have been acquired and lead to immense challenges for subsequent image analysis.
Despite the incredible progress made in the field of neuroimage informatics, a variety of problems still remain, related, for instance, to the constant desire of increasing the accuracy of neuron segmentation and tracing.
Problems can emerge at every stage of the image analysis pipeline, including distributed computation, data management, pre-processing, segmentation, detection, proofreading, and qualitative and quantitative analysis. For example, part of the imaged cell membrane could be missing due to a staining issue and contacting neurons could be merged in the automated segmentation posing a challenging task to the experimenter. Correcting these errors requires additional time, effort, and human expertise, representing a potential bottleneck while reconstructing neurons within a functional circuit.
The goal of this Research Topic is to identify and tackle image processing and analysis difficulties in neural connectomics research.
We welcome all types of articles covering and providing further insights into methods and software applications helping to solve problems in the field of neural connectomics research.
These topics include:
• Computational architecture including distributed computation and cloud computing
• Functional imaging software and related video or image processing
• Structural imaging software and related image processing
• Image pre-processing including defect detection and correction
• Image alignment
• Ground truth labeling
• Image segmentation and object detection
• Visualization
• Proofreading
Benefiting from recent breakthroughs in imaging techniques, big samples could be imaged with high resolution. With light microscopy, a whole mouse or rhesus monkey brain could be imaged at the level of single neurite resolution. Electron microscopy allows on the other hand to image a cubic millimeter of brain tissue at an unparalleled nanometer resolution, consenting to resolve synapses and trace nanometer-thin neuronal processes. Other imaging methods, such as Synchrotron-Based X-ray Imaging, are improving and have already produced impressive images.
New tissue clearing systems enable efficient imaging of high-resolution individual nerve axons in the brain, spinal cord, and peripheral organs. A variety of large-scale image volumes have been acquired and lead to immense challenges for subsequent image analysis.
Despite the incredible progress made in the field of neuroimage informatics, a variety of problems still remain, related, for instance, to the constant desire of increasing the accuracy of neuron segmentation and tracing.
Problems can emerge at every stage of the image analysis pipeline, including distributed computation, data management, pre-processing, segmentation, detection, proofreading, and qualitative and quantitative analysis. For example, part of the imaged cell membrane could be missing due to a staining issue and contacting neurons could be merged in the automated segmentation posing a challenging task to the experimenter. Correcting these errors requires additional time, effort, and human expertise, representing a potential bottleneck while reconstructing neurons within a functional circuit.
The goal of this Research Topic is to identify and tackle image processing and analysis difficulties in neural connectomics research.
We welcome all types of articles covering and providing further insights into methods and software applications helping to solve problems in the field of neural connectomics research.
These topics include:
• Computational architecture including distributed computation and cloud computing
• Functional imaging software and related video or image processing
• Structural imaging software and related image processing
• Image pre-processing including defect detection and correction
• Image alignment
• Ground truth labeling
• Image segmentation and object detection
• Visualization
• Proofreading
Keywords: image processing, neuroinformatics, 3D image, deep learning, machine learning segmentation, skeletonization
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
