scholarly journals Deep learning based behavioral profiling of rodent stroke recovery

2021 ◽  
Author(s):  
Rebecca Zoe Weber ◽  
Geertje Mulders ◽  
Julia Kaiser ◽  
Christian Tackenberg ◽  
Ruslan Rust
Keyword(s):  
Deep Learning ◽  
Motion Tracking ◽  
Time Course ◽  
Focal Cerebral Ischemia ◽  
Underlying Disease ◽  
Stroke Recovery ◽  
Mouse Strains ◽  
Body Parts ◽  
3D Tracking ◽  
Stroke Research

Stroke research heavily relies on rodent behavior when assessing underlying disease mechanisms and treatment efficacy. Although functional motor recovery is considered the primary targeted outcome, tests in rodents are still poorly reproducible, and often unsuitable for unraveling the complex behavior after injury. Here, we provide a comprehensive 3D gait analysis of mice after focal cerebral ischemia based on the new deep learning-based software (DeepLabCut, DLC) that only requires basic behavioral equipment. We demonstrate a high precision 3D tracking of 10 body parts (including all relevant joints and reference landmarks) in several mouse strains with an accuracy of 99.4%. Building on this rigor motion tracking, a comprehensive post-analysis (with >100 parameters) unveils biologically relevant differences in locomotor profiles after a stroke over a time course of three weeks. We further refine the widely used ladder rung test using deep learning and compare its performance to human annotators. The generated DLC-assisted tests were then benchmarked to five widely used conventional behavioral set-ups (neurological scoring, rotarod, ladder rung walk, cylinder test, and single-pellet grasping) regarding sensitivity, accuracy, time use and costs. We conclude that deep learning-based motion tracking with comprehensive post-analysis provides accurate and sensitive data to describe the complex recovery of rodents following a stroke. The experimental set-up and analysis can also benefit a range of other neurological injuries that affect locomotion.

Spatiotemporal electrophysiology of cerebral ischemia observed using chronic electrode array in auditory cortex

2008 ◽  
Vol 2 (1) ◽  
Author(s):  
M.T. Huberty ◽  
P. Tek ◽  
P.J. Rousche
Keyword(s):  
Auditory Cortex ◽  
Time Course ◽  
Electrode Array ◽  
The United States ◽  
Stroke Recovery ◽  
Sprague Dawley ◽  
Stroke Research ◽  
Sprague Dawley Rats ◽  
The Mean

Stroke research is of considerable societal value in an age in which the scourge is a leading cause of disability and the third-leading cause of death in the United States. While previous studies investigate the electrophysiology of stroke, none examine the long-term time-course of stroke recovery in the auditory cortex, the objective of this study. An electrode was implanted in the auditory cortex of two anesthetized Sprague-Dawley rats, stroke was induced in one of the subjects using photothrombosis, and daily electrical recordings were made while each subject was presented with a click stimulus every 500 ms. Peri-stimulus time histograms reveal that in the control subject, the second stimulus-evoked bursts peak decreased the day following implantation (Day 1) but returned almost to its Day 0 (day of surgery) value by Day 5, representing recovery from implantation trauma. The mean firing rate decreased logarithmically from its Day 0 value of 90 Hz to 10 Hz by Day 8, revealing decreasing electrode viability. In the stroke subject, the second stimulus-evoked bursts peak was undetected Day 1, but was detected again on Day 4, elucidating that the rat auditory cortex regains function as stroke recovery progresses.

scholarly journals Assessment of patient specific information in the wild on fundus photography and optical coherence tomography

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Marion R. Munk ◽  
Thomas Kurmann ◽  
Pablo Márquez-Neila ◽  
Martin S. Zinkernagel ◽  
Sebastian Wolf ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cross Sections ◽  
Underlying Disease ◽  
Geographic Atrophy ◽  
Clinical Population ◽  
Fundus Photography ◽  
Peak Performance ◽  
Patient Specific ◽  
Specific Information ◽  
Fundus Images

AbstractIn this paper we analyse the performance of machine learning methods in predicting patient information such as age or sex solely from retinal imaging modalities in a heterogeneous clinical population. Our dataset consists of N = 135,667 fundus images and N = 85,536 volumetric OCT scans. Deep learning models were trained to predict the patient’s age and sex from fundus images, OCT cross sections and OCT volumes. For sex prediction, a ROC AUC of 0.80 was achieved for fundus images, 0.84 for OCT cross sections and 0.90 for OCT volumes. Age prediction mean absolute errors of 6.328 years for fundus, 5.625 years for OCT cross sections and 4.541 for OCT volumes were observed. We assess the performance of OCT scans containing different biomarkers and note a peak performance of AUC = 0.88 for OCT cross sections and 0.95 for volumes when there is no pathology on scans. Performance drops in case of drusen, fibrovascular pigment epitheliuum detachment and geographic atrophy present. We conclude that deep learning based methods are capable of classifying the patient’s sex and age from color fundus photography and OCT for a broad spectrum of patients irrespective of underlying disease or image quality. Non-random sex prediction using fundus images seems only possible if the eye fovea and optic disc are visible.

scholarly journals Deep learning-based pupil model predicts time and spectral dependent light responses

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Babak Zandi ◽  
Tran Quoc Khanh
Keyword(s):  
Deep Learning ◽  
Time Course ◽  
Light Response ◽  
Absolute Error ◽  
Light Responses ◽  
Temporal Prediction ◽  
Feed Forward Neural Networks ◽  
Self Learning ◽  
Pupil Light Response ◽  
Planckian Locus

AbstractAlthough research has made significant findings in the neurophysiological process behind the pupillary light reflex, the temporal prediction of the pupil diameter triggered by polychromatic or chromatic stimulus spectra is still not possible. State of the art pupil models rested in estimating a static diameter at the equilibrium-state for spectra along the Planckian locus. Neither the temporal receptor-weighting nor the spectral-dependent adaptation behaviour of the afferent pupil control path is mapped in such functions. Here we propose a deep learning-driven concept of a pupil model, which reconstructs the pupil’s time course either from photometric and colourimetric or receptor-based stimulus quantities. By merging feed-forward neural networks with a biomechanical differential equation, we predict the temporal pupil light response with a mean absolute error below 0.1 mm from polychromatic (2007 $$\pm$$ ± 1 K, 4983 $$\pm$$ ± 3 K, 10,138 $$\pm$$ ± 22 K) and chromatic spectra (450 nm, 530 nm, 610 nm, 660 nm) at 100.01 ± 0.25 cd/m2. This non-parametric and self-learning concept could open the door to a generalized description of the pupil behaviour.

scholarly journals A Protective Role for Interleukin-1 Signaling during Mouse Adenovirus Type 1-Induced Encephalitis

2016 ◽  
Vol 91 (4) ◽  
Author(s):  
Luiza A. Castro-Jorge ◽  
Carla D. Pretto ◽  
Asa B. Smith ◽  
Oded Foreman ◽  
Kelly E. Carnahan ◽  
...  
Keyword(s):  
Inflammatory Cytokine ◽  
Time Course ◽  
Interleukin 1 ◽  
Adenovirus Type ◽  
Mouse Strains ◽  
Complex Nature ◽  
Type I ◽  
Viral Loads ◽  
The Brain

ABSTRACT Interleukin-1β (IL-1β), an inflammatory cytokine and IL-1 receptor ligand, has diverse activities in the brain. We examined whether IL-1 signaling contributes to the encephalitis observed in mouse adenovirus type 1 (MAV-1) infection, using mice lacking the IL-1 receptor (Il1r1 −/− mice). Il1r1 −/− mice demonstrated reduced survival, greater disruption of the blood-brain barrier (BBB), higher brain viral loads, and higher brain inflammatory cytokine and chemokine levels than control C57BL/6J mice. We also examined infections of mice defective in IL-1β production (Pycard −/− mice) and mice defective in trafficking of Toll-like receptors to the endosome (Unc93b1 −/− mice). Pycard −/− and Unc93b1 −/− mice showed lower survival (similar to Il1r1 −/− mice) than control mice but, unlike Il1r1 −/− mice, did not have increased brain viral loads or BBB disruption. Based on the brain cytokine levels, MAV-1-infected Unc93b1 −/− mice had a very different inflammatory profile from infected Il1r1 −/− and Pycard −/− mice. Histological examination demonstrated pathological findings consistent with encephalitis in control and knockout mice; however, intranuclear viral inclusions were seen only in Il1r1 −/− mice. A time course of infection of control and Il1r1 −/− mice evaluating the kinetics of viral replication and cytokine production revealed differences between the mouse strains primarily at 7 to 8 days after infection, when mice began succumbing to MAV-1 infection. In the absence of IL-1 signaling, we noted an increase in the transcription of type I interferon (IFN)-stimulated genes. Together, these results indicate that IL-1 signaling is important during MAV-1 infection and suggest that, in its absence, increased IFN-β signaling may result in increased neuroinflammation. IMPORTANCE The investigation of encephalitis pathogenesis produced by different viruses is needed to characterize virus and host-specific factors that contribute to disease. MAV-1 produces viral encephalitis in its natural host, providing a good model for studying factors involved in encephalitis development. We investigated the role of IL-1 signaling during MAV-1-induced encephalitis. Unexpectedly, the lack of IL-1 signaling increased the mortality and inflammation in mice infected with MAV-1. Also, there was an increase in the transcription of type I IFN-stimulated genes that correlated with the observed increased mortality and inflammation. The findings highlight the complex nature of encephalitis and suggests that IL-1 has a protective effect for the development of MAV-1-induced encephalitis.

scholarly journals Immunopathological Basis of Lymphocytic Choriomeningitis Virus-Induced Chorioretinitis and Keratitis

2008 ◽  
Vol 83 (1) ◽  
pp. 159-166 ◽  
Author(s):  
Martin S. Zinkernagel ◽  
Beatrice Bolinger ◽  
Philippe Krebs ◽  
Lucas Onder ◽  
Simone Miller ◽  
...  
Keyword(s):  
T Cells ◽  
T Cell ◽  
Time Course ◽  
Immunosuppressive Agents ◽  
Passive Immunization ◽  
Central Nervous System Disease ◽  
Lymphocytic Choriomeningitis ◽  
Lymphocytic Choriomeningitis Virus ◽  
Ocular Disease ◽  
Mouse Strains

ABSTRACT The infection of humans with the rodent-borne lymphocytic choriomeningitis virus (LCMV) can lead to central nervous system disease in adults or severe neurological disease with hydrocephalus and chorioretinitis in children infected congenitally. Although LCMV-induced meningitis and encephalitis have been studied extensively, the immunopathological mechanisms underlying LCMV infection-associated ocular disease remain elusive. We report here that the intraocular administration of the neurotropic LCMV strain Armstrong (Arm) elicited pronounced chorioretinitis and keratitis and that infection with the more viscerotropic strains WE and Docile precipitated less severe immunopathological ocular disease. Time course analyses revealed that LCMV Arm infection of the uvea and neuroretina led to monophasic chorioretinitis which peaked between days 7 and 12 after infection. Analyses of T-cell-deficient mouse strains showed that LCMV-mediated ocular disease was strictly dependent on the presence of virus-specific CD8+ T cells and that the contribution of CD4+ T cells was negligible. Whereas the topical application of immunosuppressive agents did not prevent the development of chorioretinitis, passive immunization with hyperimmune sera partially prevented retinal and corneal damage. Likewise, mice displaying preexisting LCMV-specific T-cell responses were protected against LCMV-induced ocular disease. Thus, antibody- and/or T-cell-based vaccination protocols could be employed as preventive strategies against LCMV-mediated chorioretinitis.

scholarly journals Effects of Nitrofurantoin on the Primary and Secondary Reproductive Organs of Female B6C3F Mice

1989 ◽  
Vol 17 (4_part_2) ◽  
pp. 774-781 ◽  
Author(s):  
Katherine A. Stitzel ◽  
R. Frank McConnell ◽  
T. A. Dierckman
Keyword(s):  
Body Weight ◽  
Time Course ◽  
Body Weight Gain ◽  
Reproductive Organs ◽  
Ovarian Tumors ◽  
Feed Consumption ◽  
Mouse Strains ◽  
Carcinogenicity Study ◽  
High Doses ◽  
Benign Ovarian Tumors

The National Toxicology Program (NTP) has reported female mice fed high doses of Nitrofurantoin (NFT) were found to have ovarian atrophy as diagnosed histologically and increased benign ovarian tumors after 24 months of exposure (30). This result contrasts with 4 other recent carcinogenicity assays in rodents with NFT, all with no evidence of an ovarian effect. An extensive database documents benign tubular adenomas develop secondary to ovarian atrophy in many mouse strains, including B6C3F1 (see ref 11). The present study was initiated to confirm this mechanism could be responsible for the ovarian tumors in the NTP study and to investigate the time course of ovarian changes seen in female B6C3F1 mice. Mice were provided diet containing NFT at doses of 350 and 500 mg/kg body weight/day and examined after 4, 8, 13, 17, 43 and 64 weeks. A dose-related decrease in feed consumption, feed efficiency and body weight gain was seen and persisted throughout the study. Sexual maturity was delayed in a dose-related fashion, compatible with previously reported effects of reduced food consumption in rodents (12, 16). All groups of mice eventually did have normal estrous cycles, but cycle lengths were increased in a dose-related fashion. Both doses of NFT resulted in histological evidence of senile ovarian atrophy by week 43. Based on the reported association between sterility and ovarian tumors, we conclude the benign tubular adenomas seen at 2 yr in the NTP carcinogenicity study with NFT were secondary to the ovarian atrophy induced in this strain of mouse and not an indication NFT, itself, is a carcinogen.

Characterization of a new double-filament model of focal cerebral ischemia in heme oxygenase-2-deficient mice

2003 ◽  
Vol 285 (1) ◽  
pp. R222-R230 ◽  
Author(s):  
Shozo Goto ◽  
Kenji Sampei ◽  
Nabil J. Alkayed ◽  
Sylvain Doré ◽  
Raymond C. Koehler
Keyword(s):  
At Risk ◽  
Blood Flow ◽  
Heme Oxygenase ◽  
Focal Cerebral Ischemia ◽  
Infarct Volume ◽  
Neuroprotective Effect ◽  
Cerebral Arteries ◽  
Vascular Anatomy ◽  
Mouse Strains ◽  
Left Internal Carotid Artery

Variations in vascular anatomy in knockout mouse strains can influence infarct volume after middle cerebral artery (MCA) occlusion (MCAO). In wild-type (WT) and heme oxygenase-2 gene-deleted (HO2-/-) mice, infarcts were not reproducibly achieved with the standard intraluminal filament technique. The present study characterizes a double-filament model of MCAO, which was developed to produce consistent infarcts in both WT and HO2-/- mice. Diameters of most cerebral arteries were similar in WT and HO2-/- mice, although the posterior communicating artery size was variable. In halothane-anesthetized mice, two 6-0 monofilaments with blunted tips were inserted into the left internal carotid artery 6.0 and 4.5 mm past the pterygopalatine artery junction to reside distal and proximal to the origin of the MCA. The tissue “volume at risk” determined by brief dye perfusion in WT (59 ± 2% of hemisphere; ±SE) was similar to HO2-/- (62 ± 4%). The volume of tissue with cerebral blood flow <50 ml·min-1·100 g-1 was similar in WT (35 ± 9%) and HO2-/- (36 ± 11%) during MCAO and at 3 h of reperfusion (<2%). After 1 h MCAO, infarct volume was greater in HO2-/- (44 ± 6%) than WT (25 ± 3%). After increasing MCAO duration to 2 h, the difference between HO2-/- (47 ± 4%) and WT (36 ± 3%) diminished, but infarct volume remained substantially less than the volume at risk. Infusion of tin protoporphyrin IX, an HO inhibitor, during reperfusion after 1 h MCAO increased infarct volume in WT but not significantly in HO2-/- mice, although infarct volume remained less than the volume at risk. Thus greater infarct volume in HO2-/- mice is not attributable to a greater volume at risk, lower intraischemic blood flow, or poor reflow, but rather to a neuroprotective effect of HO2 activity. The double-filament model may be of use as an alternative in other murine knockout strains in which the standard filament model does not yield consistent infarcts.

scholarly journals A general end-to-end diagnosis framework for manufacturing systems

2019 ◽  
Vol 7 (2) ◽  
pp. 418-429 ◽  
Author(s):  
Ye Yuan ◽  
Guijun Ma ◽  
Cheng Cheng ◽  
Beitong Zhou ◽  
Huan Zhao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Manufacturing Systems ◽  
Time Course ◽  
Manufacturing Sector ◽  
Smart Manufacturing ◽  
Data Sets ◽  
Learning Techniques ◽  
General Data ◽  
End To End ◽  
Manufacturing Applications

Abstract The manufacturing sector is envisioned to be heavily influenced by artificial-intelligence-based technologies with the extraordinary increases in computational power and data volumes. A central challenge in the manufacturing sector lies in the requirement of a general framework to ensure satisfied diagnosis and monitoring performances in different manufacturing applications. Here, we propose a general data-driven, end-to-end framework for the monitoring of manufacturing systems. This framework, derived from deep-learning techniques, evaluates fused sensory measurements to detect and even predict faults and wearing conditions. This work exploits the predictive power of deep learning to automatically extract hidden degradation features from noisy, time-course data. We have experimented the proposed framework on 10 representative data sets drawn from a wide variety of manufacturing applications. Results reveal that the framework performs well in examined benchmark applications and can be applied in diverse contexts, indicating its potential use as a critical cornerstone in smart manufacturing.

Sign in / Sign up

Export Citation Format

Share Document