Modeling spatiotemporal patterns of gait anomaly with a CNN-LSTM deep neural network

Sadeghzadehyazdi, Nasrin; Batabyal, Tamal; Acton, Scott T.

doi:10.1016/j.eswa.2021.115582

Cited by 24 publications

(9 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A gait anomaly acknowledgement is proposed by Sadeghzadehyazdi et al [25], which makes use of Kinect skeleton data to capture spatiotemporal patterns. The suggested model represents the interdependence of various body joints during locomotion by taking the skeleton as a whole into account.…”

Section: Deep Learning-based Gait Analysismentioning

confidence: 99%

Gait Based Person Identification Using Deep Learning Model of Generative Adversarial Network

Vatambeti¹,

Damera²

2022

ATAIML

View full text Add to dashboard Cite

The proliferation of digital age security tools is often attributed to the rise of visual surveillance. Since an individual's gait is highly indicative of their identity, it is becoming an increasingly popular biometric modality for use in autonomous visual surveillance and monitoring. There are various steps used in gait recognition frameworks such as segmentation, feature extraction, feature learning and similarity measurement. These steps are mutually independent with each part fixed, which results in a suboptimal performance in a challenging condition. It can be done independently of the users' involvement. Low-resolution video and straightforward instrumentation can verify an individual's identity, making impersonation a rarity. Using the benefits of the Generative Adversarial Network (GAN), this investigation tackles the problem of unevenly distributed unlabeled data with infrequently performed tasks. To estimate the data circulation in various circumstances using constrained observed gait data, a multimodal generator is applied here. When it comes to sharing knowledge, the variety provided by the data generated by a multimodal generator is hard to beat. The capability to distinguish gait activities with varying patterns due to environmental dynamics is enhanced by this multimodal generator. This system is more stable than other gait-based recognition methods because it can process data that is not equally dispersed throughout a different environment. The system's reliability is enhanced by the multimodal generator's capacity to produce a wide variety of outputs. The testing results show that this algorithm is superior to other gait-based recognition methods because it can adapt to changing environments.

show abstract

Section: Deep Learning-based Gait Analysismentioning

confidence: 99%

Gait Based Person Identification Using Deep Learning Model of Generative Adversarial Network

Vatambeti¹,

Damera²

2022

ATAIML

View full text Add to dashboard Cite

show abstract

“…CNN is good at extracting local features such as phrases, while BiLSTM can extract text context information and long-distance dependence information. The purpose of constructing this model is to make use of these two advantages [ 30 , 31 ].…”

Section: English Text Readability Measurement Based On Convolutional ...mentioning

confidence: 99%

“…Pearson correlation coefficient Model 3 [30] 0.732 -Model 4 [31] 0.628 -Model 1 [33] 0.911 e proposed model 0.775 0.836…”

Section: Model Accuracymentioning

confidence: 99%

English Text Readability Measurement Based on Convolutional Neural Network: A Hybrid Network Model

Jian

Xiang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Text readability is very important in meeting people’s information needs. With the explosive growth of modern information, the measurement demand of text readability is increasing. In view of the text structure of words, sentences, and texts, a hybrid network model based on convolutional neural network is proposed to measure the readability of English texts. The traditional method of English text readability measurement relies too much on the experience of artificial experts to extract features, which limits its practicability. With the increasing variety and quantity of text readability measurement features to be extracted, it is more and more difficult to extract deep features manually, and it is easy to introduce irrelevant features or redundant features, resulting in the decline of model performance. This paper introduces the concept of hybrid network model in deep learning; constructs a hybrid network model suitable for English text readability measurement by combining convolutional neural network, bidirectional long short-term memory network, and attention mechanism network; and replaces manual automatic feature extraction by machine learning, which greatly improves the measurement efficiency and performance of text readability.

show abstract

“…Deep learning-based approaches, such as convolutional neural networks (CNN), recurrent neural networks (RNN), and long short-term memory (LSTM) networks, are potential alternatives in seizure or seizure onset detection 36,37 . Deep learning configurations are superior to the state-of-the-art machine learning models in many fields, ranging from object and activity detection [38][39][40] to language modeling 41 to biological problems [42][43][44] . If properly trained with a large corpus of data, a deep learning model can extract salient features of the complex polymorphic pattern of a seizure.…”

Section: Seizure / Seizure Onset Detection Algorithmsmentioning

confidence: 99%

A deep learning-based automated closed-loop optogenetic system for neuromodulation during seizures

Batabyal

Brodovskaya

Williamson

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Closed-loop electrical brain stimulation is becoming a popular technique proposed for use as a treatment alternative to surgical resection of brain tissue for drug-resistant seizures in epilepsy patients. Closed-loop optogenetic stimulation is an experimental alternative to electrical stimulation since it can stimulate or inhibit neurons. The closed-loop part contains an online seizure detection algorithm, which, based on the design, can detect the onset of a seizure or evaluate a running window for seizure membership. Conventional configurations of closed-loop optogenetics have several limitations, ranging from the adaptability of hardware-based implementation to inadequate, customized feature selection of seizures, among others. Here we provide a detailed description of our closed-loop components. We used a sequential, fully convolutional neural network regressor for complex feature selection of seizures against the controls from local field potential recordings. Our modular design kept the local field potential-recording headset and optical probe separate. The seizure detection and execution of light delivery are fast and can be precisely timed. This automated system is robust to noise, modular in design, flexible in use, and simple execution. When applied in vivo, the proposed work shows efficacy over the state of the arts in terms of improved seizure detection and reduction in false positives.

show abstract

Modeling spatiotemporal patterns of gait anomaly with a CNN-LSTM deep neural network

Cited by 24 publications

References 43 publications

Gait Based Person Identification Using Deep Learning Model of Generative Adversarial Network

Gait Based Person Identification Using Deep Learning Model of Generative Adversarial Network

English Text Readability Measurement Based on Convolutional Neural Network: A Hybrid Network Model

A deep learning-based automated closed-loop optogenetic system for neuromodulation during seizures

Contact Info

Product

Resources

About