Browse
Submit Data
Databases
API
Help

Dataset Information

23 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

LiftPose3D, a deep learning-based approach for transforming two-dimensional to three-dimensional poses in laboratory animals.

ABSTRACT: Markerless three-dimensional (3D) pose estimation has become an indispensable tool for kinematic studies of laboratory animals. Most current methods recover 3D poses by multi-view triangulation of deep network-based two-dimensional (2D) pose estimates. However, triangulation requires multiple synchronized cameras and elaborate calibration protocols that hinder its widespread adoption in laboratory studies. Here we describe LiftPose3D, a deep network-based method that overcomes these barriers by reconstructing 3D poses from a single 2D camera view. We illustrate LiftPose3D's versatility by applying it to multiple experimental systems using flies, mice, rats and macaques, and in circumstances where 3D triangulation is impractical or impossible. Our framework achieves accurate lifting for stereotypical and nonstereotypical behaviors from different camera angles. Thus, LiftPose3D permits high-quality 3D pose estimation in the absence of complex camera arrays and tedious calibration procedures and despite occluded body parts in freely behaving animals.

SUBMITTER: Gosztolai A

PROVIDER: S-EPMC7611544 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Similar Datasets

DeepBhvTracking: A Novel Behavior Tracking Method for Laboratory Animals Based on Deep Learning.

Project description:Behavioral measurement and evaluation are broadly used to understand brain functions in neuroscience, especially for investigations of movement disorders, social deficits, and mental diseases. Numerous commercial software and open-source programs have been developed for tracking the movement of laboratory animals, allowing animal behavior to be analyzed digitally. In vivo optical imaging and electrophysiological recording in freely behaving animals are now widely used to understand neural functions in circuits. However, it is always a challenge to accurately track the movement of an animal under certain complex conditions due to uneven environment illumination, variations in animal models, and interference from recording devices and experimenters. To overcome these challenges, we have developed a strategy to track the movement of an animal by combining a deep learning technique, the You Only Look Once (YOLO) algorithm, with a background subtraction algorithm, a method we label DeepBhvTracking. In our method, we first train the detector using manually labeled images and a pretrained deep-learning neural network combined with YOLO, then generate bounding boxes of the targets using the trained detector, and finally track the center of the targets by calculating their centroid in the bounding box using background subtraction. Using DeepBhvTracking, the movement of animals can be tracked accurately in complex environments and can be used in different behavior paradigms and for different animal models. Therefore, DeepBhvTracking can be broadly used in studies of neuroscience, medicine, and machine learning algorithms.

| S-EPMC8581673 | biostudies-literature

Quantifying two-dimensional and three-dimensional stereoscopic learning in anatomy using electroencephalography.

Project description:Advances in computer visualization enabling both 2D and 3D representation have generated tools to aid perception of spatial relationships and provide a new forum for instructional design. A key knowledge gap is the lack of understanding of how the brain neurobiologically processes and learns from spatially presented content, and new quantitative variables are required to address this gap. The objective of this study was to apply quantitative neural measures derived from electroencephalography (EEG) to examine stereopsis in anatomy learning by comparing mean amplitude changes in N250 (related to object recognition) and reward positivity (related to responding to feedback) event related to potential components using a reinforcement-based learning paradigm. Health sciences students (n = 61) learned to identify and localize neuroanatomical structures using 2D, 3D, or a combination of models while EEG and behavioral (accuracy) data were recorded. Participants learning using 3D models had a greater object recognition (N250 amplitude) compared to those who learned from 2D models. Based on neurological results, interleaved learning incorporating both 2D and 3D models provided an advantage in learning, retention, and transfer activities represented by decreased reward positivity amplitude. Behavioral data did not have the same sensitivity as neural data for distinguishing differences in learning with and without stereopsis in these learning activities. Measuring neural activity reveals new insights in applied settings for educators to consider when incorporating stereoscopic models in the design of learning interventions.

| S-EPMC6637108 | biostudies-literature

Recording Neural Activity in Unrestrained Animals with Three-Dimensional Tracking Two-Photon Microscopy.

Project description:Optical recordings of neural activity in behaving animals can reveal the neural correlates of decision making, but brain motion, which often accompanies behavior, compromises these measurements. Two-photon point-scanning microscopy is especially sensitive to motion artifacts, and two-photon recording of activity has required rigid coupling between the brain and microscope. We developed a two-photon tracking microscope with extremely low-latency (360 μs) feedback implemented in hardware. This microscope can maintain continuous focus on neurons moving with velocities of 3 mm/s and accelerations of 1 m/s2 both in-plane and axially. We recorded calcium dynamics of motor neurons and inter-neurons in unrestrained freely behaving fruit fly larvae, correlating neural activity with stimulus presentations and behavioral outputs, and we measured light-induced depolarization of a visual interneuron in a moving animal using a genetically encoded voltage indicator. Our technique can be extended to stabilize recordings in a variety of moving substrates.

| S-EPMC6287944 | biostudies-literature

Response of fibroblasts to transforming growth factor-?1 on two-dimensional and in three-dimensional hyaluronan hydrogels.

Project description:Transforming growth factor-?1 (TGF-?1), an important cytokine with multiple functions, is secreted during wound healing. Previous studies have utilized two-dimensional (2D) cell culture to elucidate the functions of TGF-?1; however, 2D culture does not represent the complex three-dimensional (3D) in vivo environment. Using a synthetic hyaluronan (HA) extracellular matrix (ECM) hydrogel, we investigated the effect of TGF-?1 on fibroblasts cultured in three conditions--on tissue culture polystyrene (TCP), on HA (2D), and in HA (3D). After TGF-?1 treatment (0.1 to 20?ng/mL), morphological features and ECM regulation were analyzed by immunocytochemistry, Western blot, quantitative polymerase chain reaction, and zymogram assays. On TCP, cells showed the typical spindle shape with strong alpha smooth muscle actin (?-SMA) staining of cytoplasmic myofilaments along the cell axes after TGF-?1 treatment; on HA (2D), spindle-shape cells showed little ?-SMA staining; in HA (3D), cells were smaller and rounded with less ?-SMA deposition. The ?-SMA gene and protein expression on TCP were significantly upregulated by TGF-?1, but TGF-?1 did not induce ?-SMA expression in the presence of HA (both 2D and 3D). 3D HA culture significantly downregulated collagen I, III, and fibronectin expression, increased matrix metalloproteinase 1 and 2 (MMP1/MMP2) activity, upregulated MMP1 mRNA and downregulated TIMP3 mRNA expression. This study suggested that exogenous HA, particularly in 3D culture, appears to suppress ECM production, enhances ECM degradation and remodeling, and inhibits myofibroblast differentiation without decreasing TGF-? receptor expression.

| S-EPMC3501119 | biostudies-literature

A review of deep learning-based three-dimensional medical image registration methods.

Project description:Medical image registration is a vital component of many medical procedures, such as image-guided radiotherapy (IGRT), as it allows for more accurate dose-delivery and better management of side effects. Recently, the successful implementation of deep learning (DL) in various fields has prompted many research groups to apply DL to three-dimensional (3D) medical image registration. Several of these efforts have led to promising results. This review summarized the progress made in DL-based 3D image registration over the past 5 years and identify existing challenges and potential avenues for further research. The collected studies were statistically analyzed based on the region of interest (ROI), image modality, supervision method, and registration evaluation metrics. The studies were classified into three categories: deep iterative registration, supervised registration, and unsupervised registration. The studies are thoroughly reviewed and their unique contributions are highlighted. A summary is presented following a review of each category of study, discussing its advantages, challenges, and trends. Finally, the common challenges for all categories are discussed, and potential future research topics are identified.

| S-EPMC8611468 | biostudies-literature

Deep-learning approach to identifying cancer subtypes using high-dimensional genomic data.

Project description:MotivationCancer subtype classification has the potential to significantly improve disease prognosis and develop individualized patient management. Existing methods are limited by their ability to handle extremely high-dimensional data and by the influence of misleading, irrelevant factors, resulting in ambiguous and overlapping subtypes.ResultsTo address the above issues, we proposed a novel approach to disentangling and eliminating irrelevant factors by leveraging the power of deep learning. Specifically, we designed a deep-learning framework, referred to as DeepType, that performs joint supervised classification, unsupervised clustering and dimensionality reduction to learn cancer-relevant data representation with cluster structure. We applied DeepType to the METABRIC breast cancer dataset and compared its performance to state-of-the-art methods. DeepType significantly outperformed the existing methods, identifying more robust subtypes while using fewer genes. The new approach provides a framework for the derivation of more accurate and robust molecular cancer subtypes by using increasingly complex, multi-source data.Availability and implementationAn open-source software package for the proposed method is freely available at http://www.acsu.buffalo.edu/~yijunsun/lab/DeepType.html.Supplementary informationSupplementary data are available at Bioinformatics online.

| S-EPMC8215925 | biostudies-literature

A Deep Learning Approach For Programmable RNA Switches

Project description:Engineered RNA elements are programmable tools capable of detecting small molecules, proteins, and nucleic acids. Predicting the behavior of these tools remains a challenge, a situation that could be addressed through enhanced pattern recognition from deep learning. Thus, we investigate Deep Neural Networks (DNN) to predict toehold switch function as a canonical riboswitch model in synthetic biology. To facilitate DNN training, we synthesized and characterized in vivo a dataset of 91,534 toehold switches spanning 23 viral genomes and 906 human transcription factors. DNNs trained on nucleotide sequences outperformed (R2=0.43-0.70) previous state-of-the-art thermodynamic and kinetic models (R2=0.04-0.15) and allowed for human-understandable attention-visualizations (VIS4Map) to identify success and failure modes. This deep learning approach constitutes a major step forward in engineering and understanding of RNA synthetic biology.

2020-08-12 | GSE149225 | GEO

Deep learning approach for cancer subtype classification using high-dimensional gene expression data.

Project description:MotivationStudies have shown that classifying cancer subtypes can provide valuable information for a range of cancer research, from aetiology and tumour biology to prognosis and personalized treatment. Current methods usually adopt gene expression data to perform cancer subtype classification. However, cancer samples are scarce, and the high-dimensional features of their gene expression data are too sparse to allow most methods to achieve desirable classification results.ResultsIn this paper, we propose a deep learning approach by combining a convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU): our approach, DCGN, aims to achieve nonlinear dimensionality reduction and learn features to eliminate irrelevant factors in gene expression data. Specifically, DCGN first uses the synthetic minority oversampling technique algorithm to equalize data. The CNN can handle high-dimensional data without stress and extract important local features, and the BiGRU can analyse deep features and retain their important information; the DCGN captures key features by combining both neural networks to overcome the challenges of small sample sizes and sparse, high-dimensional features. In the experiments, we compared the DCGN to seven other cancer subtype classification methods using breast and bladder cancer gene expression datasets. The experimental results show that the DCGN performs better than the other seven methods and can provide more satisfactory classification results.

| S-EPMC9575247 | biostudies-literature

Deep learning methods hold promise for light fluence compensation in three-dimensional optoacoustic imaging.

Project description:SignificanceQuantitative optoacoustic imaging (QOAI) continues to be a challenge due to the influence of nonlinear optical fluence distribution, which distorts the optoacoustic image representation. Nonlinear optical fluence correction in OA imaging is highly ill-posed, leading to the inaccurate recovery of optical absorption maps. This work aims to recover the optical absorption maps using deep learning (DL) approach by correcting for the fluence effect.AimDifferent DL models were compared and investigated to enable optical absorption coefficient recovery at a particular wavelength in a nonhomogeneous foreground and background medium.ApproachData-driven models were trained with two-dimensional (2D) Blood vessel and three-dimensional (3D) numerical breast phantom with highly heterogeneous/realistic structures to correct for the nonlinear optical fluence distribution. The trained DL models such as U-Net, Fully Dense (FD) U-Net, Y-Net, FD Y-Net, Deep residual U-Net (Deep ResU-Net), and generative adversarial network (GAN) were tested to evaluate the performance of optical absorption coefficient recovery (or fluence compensation) with in-silico and in-vivo datasets.ResultsThe results indicated that FD U-Net-based deconvolution improves by about 10% over reconstructed optoacoustic images in terms of peak-signal-to-noise ratio. Further, it was observed that DL models can indeed highlight deep-seated structures with higher contrast due to fluence compensation. Importantly, the DL models were found to be about 17 times faster than solving diffusion equation for fluence correction.ConclusionsThe DL methods were able to compensate for nonlinear optical fluence distribution more effectively and improve the optoacoustic image quality.

| S-EPMC9547608 | biostudies-literature

A webcam-based machine learning approach for three-dimensional range of motion evaluation

Project description: Not available

| S-EPMC10593217 | biostudies-literature