scholarly journals Transferring and Generalizing Deep-Learning-based Neural Encoding Models across Subjects

2017 ◽  
Author(s):  
Haiguang Wen ◽  
Junxing Shi ◽  
Wei Chen ◽  
Zhongming Liu
Keyword(s):  
Deep Learning ◽  
Learning Models ◽  
Neural Encoding ◽  
Natural Vision ◽  
Cortical Responses ◽  
Single Subjects ◽  
Visual Cortical ◽  
The Relationship ◽  
Cortical Representations ◽  
The Brain

Recent studies have shown the value of using deep learning models for mapping and characterizing how the brain represents and organizes information for natural vision. However, modeling the relationship between deep learning models and the brain (or encoding models), requires measuring cortical responses to large and diverse sets of natural visual stimuli from single subjects. This requirement limits prior studies to few subjects, making it difficult to generalize findings across subjects or for a population. In this study, we developed new methods to transfer and generalize encoding models across subjects. To train encoding models specific to a subject, the models trained for other subjects were used as the prior models and were refined efficiently using Bayesian inference with a limited amount of data from the specific subject. To train encoding models for a population, the models were progressively trained and updated with incremental data from different subjects. For the proof of principle, we applied these methods to functional magnetic resonance imaging (fMRI) data from three subjects watching tens of hours of naturalistic videos, while deep residual neural network driven by image recognition was used to model the visual cortical processing. Results demonstrate that the methods developed herein provide an efficient and effective strategy to establish subject-specific or populationwide predictive models of cortical representations of high-dimensional and hierarchical visual features.

scholarly journals Domain randomization-enhanced deep learning models for bird detection

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Xin Mao ◽  
Jun Kang Chow ◽  
Pin Siang Tan ◽  
Kuan-fu Liu ◽  
Jimmy Wu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Continuous Monitoring ◽  
Bird Species ◽  
Training Data ◽  
Learning Models ◽  
Fine Grained ◽  
Bird Detection ◽  
Relationship Of ◽  
The Relationship

AbstractAutomatic bird detection in ornithological analyses is limited by the accuracy of existing models, due to the lack of training data and the difficulties in extracting the fine-grained features required to distinguish bird species. Here we apply the domain randomization strategy to enhance the accuracy of the deep learning models in bird detection. Trained with virtual birds of sufficient variations in different environments, the model tends to focus on the fine-grained features of birds and achieves higher accuracies. Based on the 100 terabytes of 2-month continuous monitoring data of egrets, our results cover the findings using conventional manual observations, e.g., vertical stratification of egrets according to body size, and also open up opportunities of long-term bird surveys requiring intensive monitoring that is impractical using conventional methods, e.g., the weather influences on egrets, and the relationship of the migration schedules between the great egrets and little egrets.

Context-Dependent Modulation of Early Visual Cortical Responses to Numerical and Nonnumerical Magnitudes

2021 ◽  
pp. 1-12
Author(s):  
Joonkoo Park ◽  
Sonia Godbole ◽  
Marty G. Woldorff ◽  
Elizabeth M. Brannon
Keyword(s):  
Visual Processing ◽  
Context Dependency ◽  
Numerical Magnitude ◽  
Continuous Variables ◽  
Processing Stream ◽  
Cortical Responses ◽  
Early Visual Cortex ◽  
Context Dependent ◽  
Visual Cortical ◽  
The Brain

Abstract Whether and how the brain encodes discrete numerical magnitude differently from continuous nonnumerical magnitude is hotly debated. In a previous set of studies, we orthogonally varied numerical (numerosity) and nonnumerical (size and spacing) dimensions of dot arrays and demonstrated a strong modulation of early visual evoked potentials (VEPs) by numerosity and not by nonnumerical dimensions. Although very little is known about the brain's response to systematic changes in continuous dimensions of a dot array, some authors intuit that the visual processing stream must be more sensitive to continuous magnitude information than to numerosity. To address this possibility, we measured VEPs of participants viewing dot arrays that changed exclusively in one nonnumerical magnitude dimension at a time (size or spacing) while holding numerosity constant and compared this to a condition where numerosity was changed while holding size and spacing constant. We found reliable but small neural sensitivity to exclusive changes in size and spacing; however, changing numerosity elicited a much more robust modulation of the VEPs. Together with previous work, these findings suggest that sensitivity to magnitude dimensions in early visual cortex is context dependent: The brain is moderately sensitive to changes in size and spacing when numerosity is held constant, but sensitivity to these continuous variables diminishes to a negligible level when numerosity is allowed to vary at the same time. Neurophysiological explanations for the encoding and context dependency of numerical and nonnumerical magnitudes are proposed within the framework of neuronal normalization.

scholarly journals Training confounder-free deep learning models for medical applications

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Qingyu Zhao ◽  
Ehsan Adeli ◽  
Kilian M. Pohl
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Bone Age ◽  
Magnetic Resonance Images ◽  
Data Sets ◽  
Large Set ◽  
Learning Models ◽  
End To End ◽  
The Impact ◽  
The Relationship

AbstractThe presence of confounding effects (or biases) is one of the most critical challenges in using deep learning to advance discovery in medical imaging studies. Confounders affect the relationship between input data (e.g., brain MRIs) and output variables (e.g., diagnosis). Improper modeling of those relationships often results in spurious and biased associations. Traditional machine learning and statistical models minimize the impact of confounders by, for example, matching data sets, stratifying data, or residualizing imaging measurements. Alternative strategies are needed for state-of-the-art deep learning models that use end-to-end training to automatically extract informative features from large set of images. In this article, we introduce an end-to-end approach for deriving features invariant to confounding factors while accounting for intrinsic correlations between the confounder(s) and prediction outcome. The method does so by exploiting concepts from traditional statistical methods and recent fair machine learning schemes. We evaluate the method on predicting the diagnosis of HIV solely from Magnetic Resonance Images (MRIs), identifying morphological sex differences in adolescence from those of the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA), and determining the bone age from X-ray images of children. The results show that our method can accurately predict while reducing biases associated with confounders. The code is available at https://github.com/qingyuzhao/br-net.

scholarly journals Automated brain tumor classification using various deep learning models: a comparative study

2021 ◽  
Vol 22 (1) ◽  
pp. 252
Author(s):  
Alaa Ahmed Abbood ◽  
Qahtan Makki Shallal ◽  
Mohammed Abdulraheem Fadhel
Keyword(s):  
Deep Learning ◽  
Brain Tumor ◽  
Brain Tumors ◽  
Comparative Study ◽  
Processing Time ◽  
Processing Unit ◽  
Automated Classification ◽  
Learning Models ◽  
Tumor Region ◽  
The Brain

The brain tumor, the most common and aggressive disease, leads to a very shorter lifespan. Thus, planning treatments is a crucial step in improving a patient's quality of life. In general, several image techniques such as CT, MRI, and ultrasound have been used for assessing tumors in the prostate, breast, lung, brain, etc. Primarily, MRI images are applied to detect tumors in the brain during this work. The enormous amount of data produced by the MRI scan thwarts tumor vs. non-tumor manual classification at a particular time. Unfortunately, with a small number of images, it has certain limitations (i.e., precise quantitative measurements). Therefore, an automated classification system is necessary to avoid human mortality. The automatic categorization of brain tumors in the surrounding tumor region is a challenging task concerning space and structural variability. Four deep learning models: AlexNet, VGG16, GoogleNet, and RestNet50, are used in this comparative study to classify brain tumors. Based on accuracy, the results showed that RestNet50 is the best model with an accuracy of 95.8%, while AlexNet has the fast performance with a processing time of 1.2 seconds. In addition, a hardware parallel processing unit (GPU) is employed for real-time purposes, where AlexNet (the fastest model) has a processing time of only 8.3 msec.

scholarly journals How does human motor cortex regulate vocal pitch in singers?

2018 ◽  
Vol 5 (8) ◽  
pp. 172208 ◽  
Author(s):  
Michel Belyk ◽  
Yune S. Lee ◽  
Steven Brown
Keyword(s):  
Motor Cortex ◽  
Vocal Folds ◽  
Neural Encoding ◽  
Human Motor Cortex ◽  
Laryngeal Muscles ◽  
Control Rules ◽  
Vocal Pitch ◽  
Human Motor ◽  
The Relationship ◽  
The Brain

Vocal pitch is used as an important communicative device by humans, as found in the melodic dimension of both speech and song. Vocal pitch is determined by the degree of tension in the vocal folds of the larynx, which itself is influenced by complex and nonlinear interactions among the laryngeal muscles. The relationship between these muscles and vocal pitch has been described by a mathematical model in the form of a set of ‘control rules’. We searched for the biological implementation of these control rules in the larynx motor cortex of the human brain. We scanned choral singers with functional magnetic resonance imaging as they produced discrete pitches at four different levels across their vocal range. While the locations of the larynx motor activations varied across singers, the activation peaks for the four pitch levels were highly consistent within each individual singer. This result was corroborated using multi-voxel pattern analysis, which demonstrated an absence of patterned activations differentiating any pairing of pitch levels. The complex and nonlinear relationships between the multiple laryngeal muscles that control vocal pitch may obscure the neural encoding of vocal pitch in the brain.

scholarly journals Natural language inference by deep learning method

2022 ◽  
Vol 355 ◽  
pp. 03028
Author(s):  
Saihan Li ◽  
Zhijie Hu ◽  
Rong Cao
Keyword(s):  
Deep Learning ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Learning Method ◽  
Word Embeddings ◽  
Learning Models ◽  
The Relationship ◽  
Inference Task

Natural Language inference refers to the problem of determining the relationships between a premise and a hypothesis, it is an emerging area of natural language processing. The paper uses deep learning methods to complete natural language inference task. The dataset includes 3GPP dataset and SNLI dataset. Gensim library is used to get the word embeddings, there are 2 methods which are word2vec and doc2vec to map the sentence to array. 2 deep learning models DNNClassifier and Attention are implemented separately to classify the relationship between the proposals from the telecommunication area dataset. The highest accuracy of the experiment is 88% and we found that the quality of the dataset decided the upper bound of the accuracy.

scholarly journals Neural Encoding and Decoding with Deep Learning for Dynamic Natural Vision

2017 ◽  
Vol 28 (12) ◽  
pp. 4136-4160 ◽  
Author(s):  
Haiguang Wen ◽  
Junxing Shi ◽  
Yizhen Zhang ◽  
Kun-Han Lu ◽  
Jiayue Cao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Neural Encoding ◽  
Natural Vision

Ultrastructure of developing visual cortical synapses in the cat superior colliculus

1991 ◽  
Vol 49 ◽  
pp. 156-157
Author(s):  
Caroline A. Miller ◽  
Laura L. Bruce
Keyword(s):  
Superior Colliculus ◽  
Phosphate Buffer ◽  
Receptive Fields ◽  
Visual Cortical Area ◽  
Cortical Synapses ◽  
Visual Cortical ◽  
And Function ◽  
Phosphate Buffered Saline ◽  
The Brain ◽  
Cat Superior Colliculus

The first visual cortical axons arrive in the cat superior colliculus by the time of birth. Adultlike receptive fields develop slowly over several weeks following birth. The developing cortical axons go through a sequence of changes before acquiring their adultlike morphology and function. To determine how these axons interact with neurons in the colliculus, cortico-collicular axons were labeled with biocytin (an anterograde neuronal tracer) and studied with electron microscopy.Deeply anesthetized animals received 200-500 nl injections of biocytin (Sigma; 5% in phosphate buffer) in the lateral suprasylvian visual cortical area. After a 24 hr survival time, the animals were deeply anesthetized and perfused with 0.9% phosphate buffered saline followed by fixation with a solution of 1.25% glutaraldehyde and 1.0% paraformaldehyde in 0.1M phosphate buffer. The brain was sectioned transversely on a vibratome at 50 μm. The tissue was processed immediately to visualize the biocytin.

Levenshtein Augmentation Improves Performance of SMILES Based Deep-Learning Synthesis Prediction

2020 ◽  
Author(s):  
Dean Sumner ◽  
Jiazhen He ◽  
Amol Thakkar ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Deep Learning ◽  
Recurrent Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Sequence Similarity ◽  
Learning Models ◽  
Underlying Network

<p>SMILES randomization, a form of data augmentation, has previously been shown to increase the performance of deep learning models compared to non-augmented baselines. Here, we propose a novel data augmentation method we call “Levenshtein augmentation” which considers local SMILES sub-sequence similarity between reactants and their respective products when creating training pairs. The performance of Levenshtein augmentation was tested using two state of the art models - transformer and sequence-to-sequence based recurrent neural networks with attention. Levenshtein augmentation demonstrated an increase performance over non-augmented, and conventionally SMILES randomization augmented data when used for training of baseline models. Furthermore, Levenshtein augmentation seemingly results in what we define as <i>attentional gain </i>– an enhancement in the pattern recognition capabilities of the underlying network to molecular motifs.</p>

Improving the Accuracy of Protein-Ligand Binding Affinity Prediction by Deep Learning Models: Benchmark and Model

2019 ◽  
Author(s):  
Mohammad Rezaei ◽  
Yanjun Li ◽  
Xiaolin Li ◽  
Chenglong Li
Keyword(s):  
Deep Learning ◽  
Drug Design ◽  
Binding Affinity ◽  
Benchmark Dataset ◽  
Rational Drug Design ◽  
Learning Models ◽  
Structure Based Drug Design ◽  
Binding Affinity Prediction ◽  
Affinity Prediction ◽  
Rational Drug

<b>Introduction:</b> The ability to discriminate among ligands binding to the same protein target in terms of their relative binding affinity lies at the heart of structure-based drug design. Any improvement in the accuracy and reliability of binding affinity prediction methods decreases the discrepancy between experimental and computational results.<br><b>Objectives:</b> The primary objectives were to find the most relevant features affecting binding affinity prediction, least use of manual feature engineering, and improving the reliability of binding affinity prediction using efficient deep learning models by tuning the model hyperparameters.<br><b>Methods:</b> The binding site of target proteins was represented as a grid box around their bound ligand. Both binary and distance-dependent occupancies were examined for how an atom affects its neighbor voxels in this grid. A combination of different features including ANOLEA, ligand elements, and Arpeggio atom types were used to represent the input. An efficient convolutional neural network (CNN) architecture, DeepAtom, was developed, trained and tested on the PDBbind v2016 dataset. Additionally an extended benchmark dataset was compiled to train and evaluate the models.<br><b>Results: </b>The best DeepAtom model showed an improved accuracy in the binding affinity prediction on PDBbind core subset (Pearson’s R=0.83) and is better than the recent state-of-the-art models in this field. In addition when the DeepAtom model was trained on our proposed benchmark dataset, it yields higher correlation compared to the baseline which confirms the value of our model.<br><b>Conclusions:</b> The promising results for the predicted binding affinities is expected to pave the way for embedding deep learning models in virtual screening and rational drug design fields.

Sign in / Sign up

Export Citation Format

Share Document