Flexible Learning-Free Segmentation and Reconstruction of Neural Volumes

Abstract

Imaging is a dominant strategy for data collection in neuroscience, yielding stacks of images that often scale to gigabytes of data for a single experiment. Machine learning algorithms from computer vision can serve as a pair of virtual eyes that tirelessly processes these images, automatically detecting and identifying microstructures. Unlike learning methods, our Flexible Learning-free Reconstruction of Imaged Neural volumes (FLoRIN) pipeline exploits structure-specific contextual clues and requires no training. This approach generalizes across different modalities, including serially-sectioned scanning electron microscopy (sSEM) of genetically labeled and contrast enhanced processes, spectral confocal reflectance (SCoRe) microscopy, and high-energy synchrotron X-ray microtomography (μCT) of large tissue volumes. We deploy the FLoRIN pipeline on newly published and novel mouse datasets, demonstrating the high biological fidelity of the pipeline’s reconstructions. FLoRIN reconstructions are of sufficient quality for preliminary biological study, for example examining the distribution and morphology of cells or extracting single axons from functional data. Compared to existing supervised learning methods, FLoRIN is one to two orders of magnitude faster and produces high-quality reconstructions that are tolerant to noise and artifacts, as is shown qualitatively and quantitatively.

Dense neuronal reconstruction through X-ray holographic nano-tomography

“…Until now, tracing of individual neuron morphologies from X-ray image data has been only possible through sparse labeling (Fonseca et al, 2018;Mizutani et al, 2013;Ng et al, 2016;Shahbazi et al, 2018). However, our results suggest that XNH image volumes contain sufficient signal to noise and spatial resolution to resolve and reconstruct dense populations of neurons without specific labeling.…”

A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells

“…For this reason, scientists put efforts into developing semi-automated tools, as well as fully automated tools, to improve segmentation efficiency. Fully automated tools, based on machine learning 13 or state-of-theart, untrained pixel classification algorithms 14 , are being improved to be used by a larger community; nevertheless, segmentation is still far from being fully reliable, and many works are still based on manual labor, which is inefficient in terms of segmentation time but still provides complete reliability. Semi-automated tools, such as ilastik 15 , represent a better compromise, as they provide an immediate readout for the segmentation that can be corrected to some extent, although it does not provide a real proofreading framework, and can be integrated using TrakEM2 in parallel 16 .…”

Learning-Free Iris Segmentation Revisited: A First Step Toward Fast Volumetric Operation Over Video Samples

“…AUC EER Mean FPS FLoRIN [26] 0.962 0.074 37.38 ± 2.58 SegNet [6] 0.992 0.016 10.50 ± 1.14 OSIRIS [21] 0.996 0.017…”

“…FLoRIN was originally developed to meet the challenges of segmenting volumes of neural microscopy, enabling automatic discovery of non-standard structures like cells, vasculature, and myelinated axons. By incorporating volumetric data into the segmentation process through the N-Dimensional Neighborhood Thresholding (NDNT, Section 3.1) algorithm [26], FLoRIN was able to boost the signal of features of interest without requiring any machine learning. In this way, FLoRIN achieved state of the art results across a number of imaging modalities in neural microscopy.…”