Key Intrinsic Connectivity Networks for Individual Identification With Siamese Long Short-Term Memory.
Ontology highlight
ABSTRACT:
SUBMITTER: Park YH
PROVIDER: S-EPMC8249867 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Individuals differ greatly in their ability to remember the details of past events, yet little is known about the brain processes that explain such individual differences in a healthy young population. Previous research suggests that episodic memory relies on functional communication among ventral regions of the default mode network ("DMN-C") that are strongly interconnected with the medial temporal lobes. In this study, we investigated whether the intrinsic functional connectivity of the DMN-C subnetwork is related to individual differences in memory ability, examining this relationship across 243 individuals (ages 18-50 years) from the openly available Cambridge Center for Aging and Neuroscience (Cam-CAN) dataset. We first estimated each participant's whole-brain intrinsic functional brain connectivity by combining data from resting-state, movie-watching, and sensorimotor task scans to increase statistical power. We then examined whether intrinsic functional connectivity predicted performance on a narrative recall task. We found no evidence that functional connectivity of the DMN-C, with itself, with other related DMN subnetworks, or with the rest of the brain, was related to narrative recall. Exploratory connectome-based predictive modeling (CBPM) analyses of the entire connectome revealed a whole-brain multivariate pattern that predicted performance, although these changes were largely outside of known memory networks. These results add to emerging evidence suggesting that individual differences in memory cannot be easily explained by brain differences in areas typically associated with episodic memory function.
| S-EPMC10925185 | biostudies-literature
Project description:When people sense that another person tries to control their decisions, some people will act against the control, whereas others will not. This individual tendency to control-averse behavior can have far-reaching consequences, such as engagement in illegal activities or noncompliance with medical treatments. Although individual differences in control-averse behavior have been well documented in behavioral studies, their neurological basis is less well understood. Here, we use a neural trait approach to examine whether individual differences in control-averse behavior might be linked to stable brain-based characteristics. To do so, we analyze the association between intrinsic connectivity networks as measured by resting state functional magnetic resonance imaging and control-averse behavior in an economic exchange game. In this game, subjects make choices that are either free or controlled by another person, with real consequences to both interaction partners. We find that the individual level of control-averse behavior can be positively predicted by intrinsic connectivity within the salience network, but not the central executive network or the default mode network. Specifically, subjects with a more prominent connectivity hub in the dorsal anterior cingulate cortex show greater levels of control-averse behavior. This finding provides the first evidence that the heterogeneity in control-averse behavior might originate in systematic differences of the stable functional brain organization.
| S-EPMC6866545 | biostudies-literature
Project description:Decoding brain functional states underlying cognitive processes from functional MRI (fMRI) data using multivariate pattern analysis (MVPA) techniques has achieved promising performance for characterizing brain activation patterns and providing neurofeedback signals. However, it remains challenging to decode subtly distinct brain states for individual fMRI data points due to varying temporal durations and dependency among different cognitive processes. In this study, we develop a deep learning based framework for brain decoding by leveraging recent advances in intrinsic functional network modeling and sequence modeling using long short-term memory (LSTM) recurrent neural networks (RNNs). Particularly, subject-specific intrinsic functional networks (FNs) are computed from resting-state fMRI data and are used to characterize functional signals of task fMRI data with a compact representation for building brain decoding models, and LSTM RNNs are adopted to learn brain decoding mappings between functional profiles and brain states. Validation results on fMRI data from the HCP dataset have demonstrated that brain decoding models built on training data using the proposed method could learn discriminative latent feature representations and effectively distinguish subtly distinct working memory tasks of different subjects with significantly higher accuracy than conventional decoding models. Informative FNs of the brain decoding models identified as brain activation patterns of working memory tasks were largely consistent with the literature. The method also obtained promising decoding performance on motor and social cognition tasks. Our results suggest that LSTM RNNs in conjunction with FNs could build interpretable, highly accurate brain decoding models.
| S-EPMC6819260 | biostudies-literature
Project description:BackgroundRecent studies have proposed deep learning techniques, namely recurrent neural networks, to improve biomedical text mining tasks. However, these techniques rarely take advantage of existing domain-specific resources, such as ontologies. In Life and Health Sciences there is a vast and valuable set of such resources publicly available, which are continuously being updated. Biomedical ontologies are nowadays a mainstream approach to formalize existing knowledge about entities, such as genes, chemicals, phenotypes, and disorders. These resources contain supplementary information that may not be yet encoded in training data, particularly in domains with limited labeled data.ResultsWe propose a new model to detect and classify relations in text, BO-LSTM, that takes advantage of domain-specific ontologies, by representing each entity as the sequence of its ancestors in the ontology. We implemented BO-LSTM as a recurrent neural network with long short-term memory units and using open biomedical ontologies, specifically Chemical Entities of Biological Interest (ChEBI), Human Phenotype, and Gene Ontology. We assessed the performance of BO-LSTM with drug-drug interactions mentioned in a publicly available corpus from an international challenge, composed of 792 drug descriptions and 233 scientific abstracts. By using the domain-specific ontology in addition to word embeddings and WordNet, BO-LSTM improved the F1-score of both the detection and classification of drug-drug interactions, particularly in a document set with a limited number of annotations. We adapted an existing DDI extraction model with our ontology-based method, obtaining a higher F1 score than the original model. Furthermore, we developed and made available a corpus of 228 abstracts annotated with relations between genes and phenotypes, and demonstrated how BO-LSTM can be applied to other types of relations.ConclusionsOur findings demonstrate that besides the high performance of current deep learning techniques, domain-specific ontologies can still be useful to mitigate the lack of labeled data.
| S-EPMC6323831 | biostudies-literature
Project description:Relation extraction is an important topic in information extraction, as it is used to create large-scale knowledge graphs for a variety of downstream applications. Its goal is to find and extract semantic links between entity pairs in natural language sentences. Deep learning has substantially advanced neural relation extraction, allowing for the autonomous learning of semantic features. We offer an effective Chinese relation extraction model that uses bidirectional LSTM (Bi-LSTM) and an attention mechanism to extract crucial semantic information from phrases without relying on domain knowledge from lexical resources or language systems in this study. The attention mechanism included into the Bi-LSTM network allows for automatic focus on key words. Two benchmark datasets were used to create and test our models: Chinese SanWen and FinRE. The experimental results show that the SanWen dataset model outperforms the FinRE dataset model, with area under the receiver operating characteristic curve values of 0.70 and 0.50, respectively. The models trained on the SanWen and FinRE datasets achieve values of 0.44 and 0.19, respectively, for the area under the precision-recall curve. In addition, the results of repeated modeling experiments indicated that our proposed method was robust and reproducible.
| S-EPMC10495963 | biostudies-literature
Project description:The forecasting of lower limb trajectories can improve the operation of assistive devices and minimise the risk of tripping and balance loss. The aim of this work was to examine four Long Short Term Memory (LSTM) neural network architectures (Vanilla, Stacked, Bidirectional and Autoencoders) in predicting the future trajectories of lower limb kinematics, i.e. Angular Velocity (AV) and Linear Acceleration (LA). Kinematics data of foot, shank and thigh (LA and AV) were collected from 13 male and 3 female participants (28 ± 4 years old, 1.72 ± 0.07 m in height, 66 ± 10 kg in mass) who walked for 10 minutes at preferred walking speed (4.34 ± 0.43 km.h-1) and at an imposed speed (5km.h-1, 15.4% ± 7.6% faster) on a 0% gradient treadmill. The sliding window technique was adopted for training and testing the LSTM models with total kinematics time-series data of 10,500 strides. Results based on leave-one-out cross validation, suggested that the LSTM autoencoders is the top predictor of the lower limb kinematics trajectories (i.e. up to 0.1s). The normalised mean squared error was evaluated on trajectory predictions at each time-step and it obtained 2.82-5.31% for the LSTM autoencoders. The ability to predict future lower limb motions may have a wide range of applications including the design and control of bionics allowing improved human-machine interface and mitigating the risk of falls and balance loss.
| S-EPMC8341582 | biostudies-literature
Project description:Modeling metal additive manufacturing processes is of great importance because it allows for the production of objects that are closer to the desired geometry and mechanical properties. Over-deposition often takes place during laser metal deposition, especially when the deposition head changes its direction and results in more material being melted onto the substrate. Modeling over-deposition is one of the necessary steps toward online process control, as a good model can be used in a closed-loop system to adjust the deposition parameters in real-time to reduce this phenomenon. In this study, we present a long-short memory neural network to model over-deposition. The model has been trained on simple geometries such as straight tracks, spiral and V-tracks made of Inconel 718. The model shows good generalization capabilities and can predict the height of more complex and previously unseen random tracks with limited performance loss. After the addition to the training dataset of a small amount of data coming from the random tracks, the performance of the model for such additional shapes improves significantly, making this approach feasible for more general applications as well.
| S-EPMC10320393 | biostudies-literature
Project description:Certain organizational features of brain networks present in the individual are lost when central tendencies are examined in the group. Here we investigated the detailed network organization of four individuals each scanned 24 times using MRI. We discovered that the distributed network known as the default network is comprised of two separate networks possessing adjacent regions in eight or more cortical zones. A distinction between the networks is that one is coupled to the hippocampal formation while the other is not. Further exploration revealed that these two networks were juxtaposed with additional networks that themselves fractionate group-defined networks. The collective networks display a repeating spatial progression in multiple cortical zones, suggesting that they are embedded within a broad macroscale gradient. Regions contributing to the newly defined networks are spatially variable across individuals and adjacent to distinct networks, raising issues for network estimation in group-averaged data and applied endeavors, including targeted neuromodulation.
| S-EPMC5519493 | biostudies-other
Project description:Experimental techniques for measuring chromatin accessibility are expensive and time consuming, appealing for the development of computational approaches to predict open chromatin regions from DNA sequences. Along this direction, existing methods fall into two classes: one based on handcrafted k -mer features and the other based on convolutional neural networks. Although both categories have shown good performance in specific applications thus far, there still lacks a comprehensive framework to integrate useful k -mer co-occurrence information with recent advances in deep learning.We fill this gap by addressing the problem of chromatin accessibility prediction with a convolutional Long Short-Term Memory (LSTM) network with k -mer embedding. We first split DNA sequences into k -mers and pre-train k -mer embedding vectors based on the co-occurrence matrix of k -mers by using an unsupervised representation learning approach. We then construct a supervised deep learning architecture comprised of an embedding layer, three convolutional layers and a Bidirectional LSTM (BLSTM) layer for feature learning and classification. We demonstrate that our method gains high-quality fixed-length features from variable-length sequences and consistently outperforms baseline methods. We show that k -mer embedding can effectively enhance model performance by exploring different embedding strategies. We also prove the efficacy of both the convolution and the BLSTM layers by comparing two variations of the network architecture. We confirm the robustness of our model to hyper-parameters by performing sensitivity analysis. We hope our method can eventually reinforce our understanding of employing deep learning in genomic studies and shed light on research regarding mechanisms of chromatin accessibility.The source code can be downloaded from https://github.com/minxueric/ismb2017_lstm .tingchen@tsinghua.edu.cn or ruijiang@tsinghua.edu.cn.Supplementary materials are available at Bioinformatics online.
| S-EPMC5870572 | biostudies-literature
Project description:We introduce a data-driven forecasting method for high-dimensional chaotic systems using long short-term memory (LSTM) recurrent neural networks. The proposed LSTM neural networks perform inference of high-dimensional dynamical systems in their reduced order space and are shown to be an effective set of nonlinear approximators of their attractor. We demonstrate the forecasting performance of the LSTM and compare it with Gaussian processes (GPs) in time series obtained from the Lorenz 96 system, the Kuramoto-Sivashinsky equation and a prototype climate model. The LSTM networks outperform the GPs in short-term forecasting accuracy in all applications considered. A hybrid architecture, extending the LSTM with a mean stochastic model (MSM-LSTM), is proposed to ensure convergence to the invariant measure. This novel hybrid method is fully data-driven and extends the forecasting capabilities of LSTM networks.
| S-EPMC5990702 | biostudies-literature