scholarly journals Deep Learning to Decipher the Progression and Morphology of Axonal Degeneration

Cells ◽  
2021 ◽  
Vol 10 (10) ◽  
pp. 2539
Author(s):  
Alex Palumbo ◽  
Philipp Grüning ◽  
Svenja Kim Landt ◽  
Lara Eleen Heckmann ◽  
Luisa Bartram ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Cortical Neurons ◽  
Hemorrhagic Stroke ◽  
Axonal Degeneration ◽  
Axonal Swelling ◽  
Systematic Analysis ◽  
Axonal Swellings ◽  
Morphological Patterns

Axonal degeneration (AxD) is a pathological hallmark of many neurodegenerative diseases. Deciphering the morphological patterns of AxD will help to understand the underlying mechanisms and develop effective therapies. Here, we evaluated the progression of AxD in cortical neurons using a novel microfluidic device together with a deep learning tool that we developed for the enhanced-throughput analysis of AxD on microscopic images. The trained convolutional neural network (CNN) sensitively and specifically segmented the features of AxD including axons, axonal swellings, and axonal fragments. Its performance exceeded that of the human evaluators. In an in vitro model of AxD in hemorrhagic stroke induced by the hemolysis product hemin, we detected a time-dependent degeneration of axons leading to a decrease in axon area, while axonal swelling and fragment areas increased. Axonal swellings preceded axon fragmentation, suggesting that swellings may be reliable predictors of AxD. Using a recurrent neural network (RNN), we identified four morphological patterns of AxD (granular, retraction, swelling, and transport degeneration). These findings indicate a morphological heterogeneity of AxD in hemorrhagic stroke. Our EntireAxon platform enables the systematic analysis of axons and AxD in time-lapse microscopy and unravels a so-far unknown intricacy in which AxD can occur in a disease context.

scholarly journals EntireAxon: Deep learning deciphers four distinct patterns of axonal degeneration

2020 ◽  
Author(s):  
Alex Palumbo ◽  
Philipp Grüning ◽  
Svenja Kim Landt ◽  
Lara Eleen Heckmann ◽  
Luisa Bartram ◽  
...  
Keyword(s):  
Deep Learning ◽  
Microfluidic Device ◽  
Axonal Degeneration ◽  
Learning Tool ◽  
High Throughput Analysis ◽  
Systematic Analysis ◽  
Biological Condition ◽  
Morphological Heterogeneity ◽  
Axonal Swellings

AbstractDifferent axonal degeneration (AxD) patterns have been described depending on the biological condition. Until now, it remains unclear whether they are restricted to one specific condition or can occur concomitantly. Here, we present a novel microfluidic device in combination with a deep learning tool, the EntireAxon, for the high-throughput analysis of AxD. We evaluated the progression of AxD in an in vitro model of hemorrhagic stroke and observed that axonal swellings preceded axon fragmentation. We further identified four distinct morphological patterns of AxD (granular, retraction, swelling, and transport degeneration) that occur concomitantly. These findings indicate a morphological heterogeneity of AxD under pathophysiologic conditions. The newly developed microfluidic device along with the EntireAxon deep learning tool enable the systematic analysis of AxD but also unravel a so far unknown intricacy in which AxD can occur in a disease context.

scholarly journals Deep Learning Based Proarrhythmia Analysis Using Field Potentials Recorded from Human Pluripotent Stem Cells Derived Cardiomyocytes

2018 ◽  
Author(s):  
Zeinab Golgooni ◽  
Sara Mirsadeghi ◽  
Mahdieh Soleymani Baghshah ◽  
Pedram Ataee ◽  
Hossein Baharvand ◽  
...  
Keyword(s):  
Neural Network ◽  
Stem Cells ◽  
Deep Learning ◽  
Pluripotent Stem Cells ◽  
Field Potential ◽  
Input Sequence ◽  
Patient Specific ◽  
Field Potentials ◽  
Drug Induced

AbstractAimAn early characterization of drug-induced cardiotoxicity may be possible by combining comprehensive in vitro pro-arrhythmia assay and deep learning techniques. The goal of this study was to develop a deep learning method to automatically detect irregular beating rhythm as well as abnormal waveforms of field potentials in an in vitro cardiotoxicity assay using human pluripotent stem cell (hPSC) derived cardiomyocytes and multi-electrode array (MEA) system.Methods and ResultsWe included field potential waveforms from 380 experiments which obtained by application of some cardioactive drugs on healthy and/or patient-specific induced pluripotent stem cells derived cardiomyocytes (iPSC-CM). We employed convolutional and recurrent neural networks, in order to develop a new method for automatic classification of field potential recordings without using any hand-engineered features. In the proposed method, a preparation phase was initially applied to split 60-second long recordings into a series of 5-second long windows. Thereafter, the classification phase comprising of two main steps was designed. In the first step, 5-second long windows were classified using a designated convolutional neural network (CNN). In the second step, the results of 5-second long window assessments were used as the input sequence to a recurrent neural network (RNN). The output was then compared to electrophysiologist-level arrhythmia (irregularity or abnormal waveforms) detection, resulting in 0.84 accuracy, 0.84 sensitivity, 0.85 specificity, and 0.88 precision.ConclusionA novel deep learning approach based on a two-step CNN-RNN method can be used for automated analysis of “irregularity or abnormal waveforms” in an in vitro model of cardiotoxicity experiments.

scholarly journals Inhibition of the autophagic protein ULK1 attenuates axonal degeneration in vitro and in vivo, enhances translation, and modulates splicing

2020 ◽  
Vol 27 (10) ◽  
pp. 2810-2827 ◽  
Author(s):  
Björn Friedhelm Vahsen ◽  
Vinicius Toledo Ribas ◽  
Jonas Sundermeyer ◽  
Alexander Boecker ◽  
Vivian Dambeck ◽  
...  
Keyword(s):  
Cortical Neurons ◽  
Axonal Degeneration ◽  
Dominant Negative ◽  
Focal Lesion ◽  
Pathological Feature ◽  
Nerve Crush ◽  
Cord Injury ◽  
Rat Cortical Neurons

Abstract Axonal degeneration is a key and early pathological feature in traumatic and neurodegenerative disorders of the CNS. Following a focal lesion to axons, extended axonal disintegration by acute axonal degeneration (AAD) occurs within several hours. During AAD, the accumulation of autophagic proteins including Unc-51 like autophagy activating kinase 1 (ULK1) has been demonstrated, but its role is incompletely understood. Here, we study the effect of ULK1 inhibition in different models of lesion-induced axonal degeneration in vitro and in vivo. Overexpression of a dominant negative of ULK1 (ULK1.DN) in primary rat cortical neurons attenuates axotomy-induced AAD in vitro. Both ULK1.DN and the ULK1 inhibitor SBI-0206965 protect against AAD after rat optic nerve crush in vivo. ULK1.DN additionally attenuates long-term axonal degeneration after rat spinal cord injury in vivo. Mechanistically, ULK1.DN decreases autophagy and leads to an mTOR-mediated increase in translational proteins. Consistently, treatment with SBI-0206965 results in enhanced mTOR activation. ULK1.DN additionally modulates the differential splicing of the degeneration-associated genes Kif1b and Ddit3. These findings uncover ULK1 as an important mediator of axonal degeneration in vitro and in vivo, and elucidate its function in splicing, defining it as a putative therapeutic target.

An MD2-perturbing peptide has therapeutic effects in rodent and rhesus monkey models of stroke

2021 ◽  
Vol 13 (597) ◽  
pp. eabb6716
Author(s):  
Zongping Fang ◽  
Di Wu ◽  
Jiao Deng ◽  
Qianzi Yang ◽  
Xijing Zhang ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Cortical Neurons ◽  
Hemorrhagic Stroke ◽  
Rna Binding ◽  
Infarct Volume ◽  
Therapeutic Index ◽  
Therapeutic Effects ◽  
Normal Brain ◽  
Experimental Treatments

Studies have failed to translate more than 1000 experimental treatments from bench to bedside, leaving stroke as the second leading cause of death in the world. Thrombolysis within 4.5 hours is the recommended therapy for stroke and cannot be performed until neuroimaging is used to distinguish ischemic stroke from hemorrhagic stroke. Therefore, finding a common and critical therapeutic target for both ischemic and hemorrhagic stroke is appealing. Here, we report that the expression of myeloid differentiation protein 2 (MD2), which is traditionally regarded to be expressed only in microglia in the normal brain, was markedly increased in cortical neurons after stroke. We synthesized a small peptide, Trans-trans-activating (Tat)–cold-inducible RNA binding protein (Tat-CIRP), which perturbed the function of MD2 and strongly protected neurons against excitotoxic injury in vitro. In addition, systemic administration of Tat-CIRP or genetic deletion of MD2 induced robust neuroprotection against ischemic and hemorrhagic stroke in mice. Tat-CIRP reduced the brain infarct volume and preserved neurological function in rhesus monkeys 30 days after ischemic stroke. Tat-CIRP efficiently crossed the blood-brain barrier and showed a wide therapeutic index for stroke because no toxicity was detected when high doses were administered to the mice. Furthermore, we demonstrated that MD2 elicited neuronal apoptosis and necroptosis via a TLR4-independent, Sam68-related cascade. In summary, Tat-CIRP provides robust neuroprotection against stroke in rodents and gyrencephalic nonhuman primates. Further efforts should be made to translate these findings to treat both ischemic and hemorrhagic stroke in patients.

scholarly journals The necroptosis machinery mediates axonal degeneration in a model of Parkinson disease

2019 ◽  
Author(s):  
Maritza Oñate ◽  
Alejandra Catenaccio ◽  
Natalia Salvadores ◽  
Cristian Saquel ◽  
Alexis Martinez ◽  
...  
Keyword(s):  
Cortical Neurons ◽  
Axonal Degeneration ◽  
Neuronal Loss ◽  
Postmortem Brain ◽  
Death Process ◽  
Early Event ◽  
Genetic Ablation ◽  
Postmortem Brain Tissue

AbstractParkinson’s disease (PD) is the second most common neurodegenerative condition, characterized by motor impairment due to the progressive degeneration of dopaminergic neurons in the substantia nigra and depletion of dopamine release in the striatum. Accumulating evidence suggest that degeneration of axons is an early event in the disease, involving destruction programs that are independent of the survival of the cell soma. Necroptosis, a programmed cell death process, is emerging as a mediator of neuronal loss in models of neurodegenerative diseases. Here, we demonstrate activation of necroptosis in postmortem brain tissue from PD patients and in a toxin-based mouse model of the disease. Inhibition of key components of the necroptotic pathway resulted in a significant delay of 6-hydroxydopamine dependent axonal degeneration of dopaminergic and cortical neurons in vitro. Genetic ablation of necroptosis mediators MLKL and RIPK3, as well as pharmacological inhibition of RIPK1 in vivo, decreased dopaminergic neuron degeneration, improving motor performance. Together, these findings suggest that axonal degeneration in PD is mediated by the necroptosis machinery, a process here referred to as necroaxoptosis, a druggable pathway to target dopaminergic neuronal loss.

scholarly journals Deep Learning-Assisted Repurposing of Plant Compounds for Treating Vascular Calcification: An In Silico Study with Experimental Validation

2022 ◽  
Vol 2022 ◽  
pp. 1-15
Author(s):  
Chia-Ter Chao ◽  
You-Tien Tsai ◽  
Wen-Ting Lee ◽  
Hsiang-Yuan Yeh ◽  
Chih-Kang Chiang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Vascular Calcification ◽  
Heterogeneous Network ◽  
Network Architecture ◽  
Representation Learning ◽  
Local Network ◽  
Computational Screening ◽  
Plant Compounds

Background. Vascular calcification (VC) constitutes subclinical vascular burden and increases cardiovascular mortality. Effective therapeutics for VC remains to be procured. We aimed to use a deep learning-based strategy to screen and uncover plant compounds that potentially can be repurposed for managing VC. Methods. We integrated drugome, interactome, and diseasome information from Comparative Toxicogenomic Database (CTD), DrugBank, PubChem, Gene Ontology (GO), and BioGrid to analyze drug-disease associations. A deep representation learning was done using a high-level description of the local network architecture and features of the entities, followed by learning the global embeddings of nodes derived from a heterogeneous network using the graph neural network architecture and a random forest classifier established for prediction. Predicted results were tested in an in vitro VC model for validity based on the probability scores. Results. We collected 6,790 compounds with available Simplified Molecular-Input Line-Entry System (SMILES) data, 11,958 GO terms, 7,238 diseases, and 25,482 proteins, followed by local embedding vectors using an end-to-end transformer network and a node2vec algorithm and global embedding vectors learned from heterogeneous network via the graph neural network. Our algorithm conferred a good distinction between potential compounds, presenting as higher prediction scores for the compound categories with a higher potential but lower scores for other categories. Probability score-dependent selection revealed that antioxidants such as sulforaphane and daidzein were potentially effective compounds against VC, while catechin had low probability. All three compounds were validated in vitro. Conclusions. Our findings exemplify the utility of deep learning in identifying promising VC-treating plant compounds. Our model can be a quick and comprehensive computational screening tool to assist in the early drug discovery process.

Feasibility Study of Deep Learning Based Radiation Sensitivity Prediction Model Using Gene Expression Profiling

2021 ◽  
Author(s):  
Euidam Kim ◽  
Yoonsun Chung
Keyword(s):  
Neural Network ◽  
Gene Expression ◽  
Deep Learning ◽  
Prediction Model ◽  
Relative Error ◽  
Absorbed Dose ◽  
Absolute Error ◽  
Radiation Sensitivity ◽  
Learning Technology

Abstract Since radiation sensitivity prediction can be used in various field, we investigate the feasibility of an in vitro radiation sensitivity prediction model using a deep neural network. A microarray of the National Cancer Institute-60 tumor cell lines and clonogenic surviving fraction at an absorbed dose of 2 Gy values are used to predict radiation sensitivity. The prediction model is based on convolutional neural network and 6-fold cross-validation approach is applied to validate the model. Of the 174 samples, 170 (97.7%) samples show less than 10% and 4 (2.30%) show more than 10% of relative error, respectively. Through an additional validation, model accurately predict 172 out of 174 samples, representing a prediction accuracy of 98.85% under the criteria of absolute error < 0.01 or the relative error < 10%. This results demonstrate that in vitro radiation sensitivity prediction from gene expression can be carried out with the deep learning technology.

scholarly journals Performance of a deep learning based neural network in the selection of human blastocysts for implantation

eLife ◽  
2020 ◽  
Vol 9 ◽  
Author(s):  
Charles L Bormann ◽  
Manoj Kumar Kanakasabapathy ◽  
Prudhvi Thirumalaraju ◽  
Raghav Gupta ◽  
Rohan Pooniwala ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Time Lapse ◽  
Vitro Fertilization ◽  
Clinical Patient ◽  
Static Images ◽  
Transfer Order ◽  
Selection Of

Deep learning in in vitro fertilization is currently being evaluated in the development of assistive tools for the determination of transfer order and implantation potential using time-lapse data collected through expensive imaging hardware. Assistive tools and algorithms that can work with static images, however, can help in improving the access to care by enabling their use with images acquired from traditional microscopes that are available to virtually all fertility centers. Here, we evaluated the use of a deep convolutional neural network (CNN), trained using single timepoint images of embryos collected at 113 hr post-insemination, in embryo selection amongst 97 clinical patient cohorts (742 embryos) and observed an accuracy of 90% in choosing the highest quality embryo available. Furthermore, a CNN trained to assess an embryo’s implantation potential directly using a set of 97 euploid embryos capable of implantation outperformed 15 trained embryologists (75.26% vs. 67.35%, p<0.0001) from five different fertility centers.

scholarly journals New Treatment for Alzheimer’s Disease, Kamikihito, Reverses Amyloid-β-Induced Progression of Tau Phosphorylation and Axonal Atrophy

2014 ◽  
Vol 2014 ◽  
pp. 1-10 ◽  
Author(s):  
Hidetoshi Watari ◽  
Yutaka Shimada ◽  
Chihiro Tohda
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Cortical Neurons ◽  
Axonal Degeneration ◽  
Amyloid Β ◽  
Tau Phosphorylation ◽  
Axonal Atrophy ◽  
Pp2a Activity

Aims.We previously reported that kamikihito (KKT), a traditional Japanese medicine, improved memory impairment and reversed the degeneration of axons in the 5XFAD mouse model of Alzheimer’s disease (AD). However, the mechanism underlying the effects of KKT remained unknown. The aim of the present study was to investigate the mechanism by which KKT reverses the progression of axonal degeneration.Methods.Primary cultured cortical neurons were treated with amyloid beta (Aβ) fragment comprising amino acid residues (25–35) (10 μM) in anin vitroAD model. KKT (10 μg/mL) was administered to the cells before or after Aβtreatment. The effects of KKT on Aβ-induced tau phosphorylation, axonal atrophy, and protein phosphatase 2A (PP2A) activity were investigated. We also performed anin vivoassay in which KKT (500 mg/kg/day) was administered to 5XFAD mice once a day for 15 days. Cerebral cortex homogenates were used to measure PP2A activity.Results.KKT improved Aβ-induced tau phosphorylation and axonal atrophy after they had already progressed. In addition, KKT increased PP2A activityin vitroandin vivo.Conclusions.KKT reversed the progression of Aβ-induced axonal degeneration. KKT reversed axonal degeneration at least in part through its role as an exogenous PP2A stimulator.

scholarly journals Deep Learning for High-Throughput Quantification of Oligodendrocyte Ensheathment at Single-Cell Resolution

2018 ◽  
Author(s):  
Yu Kang T Xu ◽  
Daryan Chitsaz ◽  
Robert A Brown ◽  
Qiao Ling Cui ◽  
Matthew A Dabarno ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
High Throughput ◽  
Heuristic Approach ◽  
The Neural Network ◽  
Automated Imaging ◽  
Parametric Analyses ◽  
Deep Learning Neural Network ◽  
High Level

AbstractHigh-throughput quantification of oligodendrocyte (OL) myelination is a significant challenge that, if addressed, would facilitate the development of therapeutics to promote myelin protection and repair. Here, we established a quantitative high-throughput method to asses OL ensheathment in-vitro, combining nanofiber culture devices and automated imaging with a heuristic approach that informed the development of a deep learning analytic algorithm. The heuristic approach was developed by modeling general characteristics of OL ensheathments, while the deep learning neural network employed a UNet architecture with enhanced capacity to associate ensheathed segments with individual OLs. Reliably extracting multiple morphological parameters from individual cells, without heuristic approximations, mimics the high-level decision-making capacity of human researchers and improves the validity of the neural network. Experimental validation demonstrated that the deep learning approach matched the accuracy of expert-human measurements of the length and number of myelin segments per cell. The combined use of automated imaging and analysis reduces tedious manual labor while eliminating variability. The capacity of this technology to perform multi-parametric analyses at the level of individual cells permits the detection of nuanced cellular differences to accelerate the discovery of new insight into OL physiology.

Sign in / Sign up

Export Citation Format

Share Document