ED-FNN: A New Deep Learning Algorithm to Detect Percentage of the Gait Cycle for Powered Prostheses

Throughout the last decade, a whole new generation of powered transtibial prostheses and exoskeletons has been developed. However, these technologies are limited by a gait phase detection which controls the wearable device as a function of the activities of the wearer. Consequently, gait phase detection is considered to be of great importance, as achieving high detection accuracy will produce a more precise, stable, and safe rehabilitation device. In this paper, we propose a novel gait percent detection algorithm that can predict a full gait cycle discretised within a 1% interval. We called this algorithm an exponentially delayed fully connected neural network (ED-FNN). A dataset was obtained from seven healthy subjects that performed daily walking activities on the flat ground and a 15-degree slope. The signals were taken from only one inertial measurement unit (IMU) attached to the lower shank. The dataset was divided into training and validation datasets for every subject, and the mean square error (MSE) error between the model prediction and the real percentage of the gait was computed. An average MSE of 0.00522 was obtained for every subject in both training and validation sets, and an average MSE of 0.006 for the training set and 0.0116 for the validation set was obtained when combining all subjects’ signals together. Although our experiments were conducted in an offline setting, due to the forecasting capabilities of the ED-FNN, our system provides an opportunity to eliminate detection delays for real-time applications.

Download Full-text

Gait Phase Detection Based on Muscle Deformation with Static Standing-Based Calibration

Sensors ◽

10.3390/s21041081 ◽

2021 ◽

Vol 21 (4) ◽

pp. 1081

Author(s):

Tamon Miyake ◽

Shintaro Yamamoto ◽

Satoshi Hosono ◽

Satoshi Funabashi ◽

Zhengxue Cheng ◽

...

Keyword(s):

Learning Algorithm ◽

Detection System ◽

Calibration Method ◽

Detection Algorithm ◽

Video Data ◽

Detection Accuracy ◽

Phase Detection ◽

Gait Phase ◽

Logistic Regression Algorithm ◽

Short Time

Gait phase detection, which detects foot-contact and foot-off states during walking, is important for various applications, such as synchronous robotic assistance and health monitoring. Gait phase detection systems have been proposed with various wearable devices, sensing inertial, electromyography, or force myography information. In this paper, we present a novel gait phase detection system with static standing-based calibration using muscle deformation information. The gait phase detection algorithm can be calibrated within a short time using muscle deformation data by standing in several postures; it is not necessary to collect data while walking for calibration. A logistic regression algorithm is used as the machine learning algorithm, and the probability output is adjusted based on the angular velocity of the sensor. An experiment is performed with 10 subjects, and the detection accuracy of foot-contact and foot-off states is evaluated using video data for each subject. The median accuracy is approximately 90% during walking based on calibration for 60 s, which shows the feasibility of the static standing-based calibration method using muscle deformation information for foot-contact and foot-off state detection.

Download Full-text

Evaluation of Gait Phase Detection Delay Compensation Strategies to Control a Gyroscope-Controlled Functional Electrical Stimulation System During Walking

Sensors ◽

10.3390/s19112471 ◽

2019 ◽

Vol 19 (11) ◽

pp. 2471 ◽

Cited By ~ 7

Author(s):

Nicole Zahradka ◽

Ahad Behboodi ◽

Henry Wright ◽

Barry Bodt ◽

Samuel Lee

Keyword(s):

Electrical Stimulation ◽

Motion Capture ◽

Functional Electrical Stimulation ◽

Gait Cycle ◽

Detection Algorithm ◽

Measurement Unit ◽

Phase Detection ◽

System Delay ◽

Finite State ◽

Gait Phase

Functional electrical stimulation systems are used as neuroprosthetic devices in rehabilitative interventions such as gait training. Stimulator triggers, implemented to control stimulation delivery, range from open- to closed-loop controllers. Finite-state controllers trigger stimulators when specific conditions are met and utilize preset sequences of stimulation. Wearable sensors provide the necessary input to differentiate gait phases during walking and trigger stimulation. However, gait phase detection is associated with inherent system delays. In this study, five stimulator triggers designed to compensate for gait phase detection delays were tested to determine which trigger most accurately delivered stimulation at the desired times of the gait cycle. Motion capture data were collected on seven typically-developing children while walking on an instrumented treadmill. Participants wore one inertial measurement unit on each ankle and gyroscope data were streamed into the gait phase detection algorithm. Five triggers, based on gait phase detection, were used to simulate stimulation to five muscle groups, bilaterally. For each condition, stimulation signals were collected in the motion capture software via analog channels and compared to the desired timing determined by kinematic and kinetic data. Results illustrate that gait phase detection is a viable finite-state control, and appropriate system delay compensations, on average, reduce stimulation delivery delays by 6.7% of the gait cycle.

Download Full-text

Classification of Shellfish Recognition Based on Improved Faster R-CNN Framework of Deep Learning

Mathematical Problems in Engineering ◽

10.1155/2021/1966848 ◽

2021 ◽

Vol 2021 ◽

pp. 1-10

Author(s):

Yiran Feng ◽

Xueheng Tao ◽

Eung-Joo Lee

Keyword(s):

Deep Learning ◽

Learning Algorithm ◽

Experimental Tests ◽

Detection Algorithm ◽

Detection Accuracy ◽

Multiple Objects ◽

Detection Model ◽

Deep Learning Algorithm ◽

Quality Sorting

In view of the current absence of any deep learning algorithm for shellfish identification in real contexts, an improved Faster R-CNN-based detection algorithm is proposed in this paper. It achieves multiobject recognition and localization through a second-order detection network and replaces the original feature extraction module with DenseNet, which can fuse multilevel feature information, increase network depth, and avoid the disappearance of network gradients. Meanwhile, the proposal merging strategy is improved with Soft-NMS, where an attenuation function is designed to replace the conventional NMS algorithm, thereby avoiding missed detection of adjacent or overlapping objects and enhancing the network detection accuracy under multiple objects. By constructing a real contexts shellfish dataset and conducting experimental tests on a vision recognition seafood sorting robot production line, we were able to detect the features of shellfish in different scenarios, and the detection accuracy was improved by nearly 4% compared to the original detection model, achieving a better detection accuracy. This provides favorable technical support for future quality sorting of seafood using the improved Faster R-CNN-based approach.

Download Full-text

Research on improved wavelet convolutional wavelet neural networks

Applied Intelligence ◽

10.1007/s10489-020-02015-5 ◽

2020 ◽

Author(s):

Jing-Wei Liu ◽

Fang-Ling Zuo ◽

Ying-Xiao Guo ◽

Tian-Yue Li ◽

Jia-Ming Chen

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Image Classification ◽

High Performance ◽

Learning Algorithm ◽

Wavelet Neural Network ◽

Deep Learning Algorithm ◽

Main Work ◽

Good Ability ◽

Fully Connected

AbstractConvolutional neural network (CNN) is recognized as state of the art of deep learning algorithm, which has a good ability on the image classification and recognition. The problems of CNN are as follows: the precision, accuracy and efficiency of CNN are expected to be improved to satisfy the requirements of high performance. The main work is as follows: Firstly, wavelet convolutional neural network (wCNN) is proposed, where wavelet transform function is added to the convolutional layers of CNN. Secondly, wavelet convolutional wavelet neural network (wCwNN) is proposed, where fully connected neural network (FCNN) of wCNN and CNN are replaced by wavelet neural network (wNN). Thirdly, image classification experiments using CNN, wCNN and wCwNN algorithms, and comparison analysis are implemented with MNIST dataset. The effect of the improved methods are as follows: (1) Both precision and accuracy are improved. (2) The mean square error and the rate of error are reduced. (3) The complexitie of the improved algorithms is increased.

Download Full-text

Multicenter Validation of a Deep Learning Detection Algorithm for Focal Cortical Dysplasia

Neurology ◽

10.1212/wnl.0000000000012698 ◽

2021 ◽

pp. 10.1212/WNL.0000000000012698

Author(s):

Ravnoor Singh Gill ◽

Hyo-Min Lee ◽

Benoit Caldairou ◽

Seok-Jun Hong ◽

Carmen Barba ◽

...

Keyword(s):

Deep Learning ◽

Risk Stratification ◽

Learning Algorithm ◽

Focal Cortical Dysplasia ◽

Detection Algorithm ◽

Cortical Dysplasia ◽

False Positives ◽

Class Iii ◽

Deep Learning Algorithm ◽

Independent Cohort

Objective.To test the hypothesis that a multicenter-validated computer deep learning algorithm detects MRI-negative focal cortical dysplasia (FCD).Methods.We used clinically-acquired 3D T1-weighted and 3D FLAIR MRI of 148 patients (median age, 23 years [range, 2-55]; 47% female) with histologically-verified FCD at nine centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed as MRI-negative in 51% of cases, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated Bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 FCD cases (13±10 years). Applying the algorithm to 42 healthy and 89 temporal lobe epilepsy disease controls tested specificity.Results.Overall sensitivity was 93% (137/148 FCD detected) using a leave-one-site-out cross-validation, with an average of six false positives per patient. Sensitivity in MRI-negative FCD was 85%. In 73% of patients, the FCD was among the clusters with the highest confidence; in half it ranked the highest. Sensitivity in the independent cohort was 83% (19/23; average of five false positives per patient). Specificity was 89% in healthy and disease controls.Conclusions.This first multicenter-validated deep learning detection algorithm yields the highest sensitivity to date in MRI-negative FCD. By pairing predictions with risk stratification this classifier may assist clinicians to adjust hypotheses relative to other tests, increasing diagnostic confidence. Moreover, generalizability across age and MRI hardware makes this approach ideal for pre-surgical evaluation of MRI-negative epilepsy.Classification of evidence.This study provides Class III evidence that deep learning on multimodal MRI accurately identifies FCD in epilepsy patients initially diagnosed as MRI-negative.

Download Full-text

Research on Network Security Application Based on Deep Learning

CONVERTER ◽

10.17762/converter.235 ◽

2021 ◽

pp. 598-605

Author(s):

Zhao Jianchao

Keyword(s):

Deep Learning ◽

Learning Algorithm ◽

Rapid Development ◽

Internet Security ◽

Detection Accuracy ◽

Data Sets ◽

Learning Technology ◽

Data Set ◽

Deep Learning Algorithm ◽

Internet Industry

Behind the rapid development of the Internet industry, Internet security has become a hidden danger. In recent years, the outstanding performance of deep learning in classification and behavior prediction based on massive data makes people begin to study how to use deep learning technology. Therefore, this paper attempts to apply deep learning to intrusion detection to learn and classify network attacks. Aiming at the nsl-kdd data set, this paper first uses the traditional classification methods and several different deep learning algorithms for learning classification. This paper deeply analyzes the correlation among data sets, algorithm characteristics and experimental classification results, and finds out the deep learning algorithm which is relatively good at. Then, a normalized coding algorithm is proposed. The experimental results show that the algorithm can improve the detection accuracy and reduce the false alarm rate.

Download Full-text

Geographic Generalization in Airborne RGB Deep Learning Tree Detection

10.1101/790071 ◽

2019 ◽

Author(s):

Ben. G. Weinstein ◽

Sergio Marconi ◽

Stephanie A. Bohlman ◽

Alina Zare ◽

Ethan P. White

Keyword(s):

Deep Learning ◽

Learning Algorithm ◽

Forest Type ◽

Detection Algorithm ◽

Forest Types ◽

Individual Tree ◽

Tree Detection ◽

Deep Learning Algorithm ◽

Local Site ◽

Using Data

AbstractTree detection is a fundamental task in remote sensing for forestry and ecosystem ecology applications. While many individual tree segmentation algorithms have been proposed, the development and testing of these algorithms is typically site specific, with few methods evaluated against data from multiple forest types simultaneously. This makes it difficult to determine the generalization of proposed approaches, and limits tree detection at broad scales. Using data from the National Ecological Observatory Network we extend a recently developed semi-supervised deep learning algorithm to include data from a range of forest types, determine whether information from one forest can be used for tree detection in other forests, and explore the potential for building a universal tree detection algorithm. We find that the deep learning approach works well for overstory tree detection across forest conditions, outperforming conventional LIDAR-only methods in all forest types. Performance was best in open oak woodlands and worst in alpine forests. When models were fit to one forest type and used to predict another, performance generally decreased, with better performance when forests were more similar in structure. However, when models were pretrained on data from other sites and then fine-tuned using a small amount of hand-labeled data from the evaluation site, they performed similarly to local site models. Most importantly, a universal model fit to data from all sites simultaneously performed as well or better than individual models trained for each local site. This result suggests that RGB tree detection models that can be applied to a wide array of forest types at broad scales should be possible.

Download Full-text

Massive MIMO Adaptive Modulation and Coding Using Online Deep Learning

10.36227/techrxiv.14526840.v2 ◽

2021 ◽

Author(s):

Evgeny Bobrov ◽

Dmitry Kropotov ◽

Hao Lu ◽

Danila Zaev

Keyword(s):

Deep Learning ◽

Massive Mimo ◽

Learning Algorithm ◽

Adaptive Modulation ◽

System Level ◽

Adaptive Modulation And Coding ◽

Q Learning ◽

Deep Learning Algorithm ◽

System Level Simulation ◽

Fully Connected

The paper describes an online deep learning algorithm for the adaptive modulation and coding in 5G Massive MIMO. The algorithm is based on a fully connected neural network, which is initially trained on the output of the traditional algorithm and then is incrementally retrained by the service feedback of its output. We show the advantage of our solution over the state-of-the-art Q-Learning approach. We provide system-level simulation results to support this conclusion in various scenarios with different channel characteristics and different user speeds. Compared with traditional OLLA our algorithm shows 10% to 20% improvement of user throughput in full buffer case. <br>

Download Full-text

IMU-Based Gait Phase Recognition for Stroke Survivors

Robotica ◽

10.1017/s0263574719000328 ◽

2019 ◽

Vol 37 (12) ◽

pp. 2195-2208 ◽

Cited By ~ 2

Author(s):

Yu Lou ◽

Rongli Wang ◽

Jingeng Mai ◽

Ninghua Wang ◽

Qining Wang

Keyword(s):

Inertial Measurement Unit ◽

Detection System ◽

Effective Means ◽

Human Motion ◽

Measurement Unit ◽

Stroke Survivors ◽

Phase Detection ◽

Wearable Robots ◽

Gait Phase ◽

Phase Recognition

SummaryUsing wearable robots is an effective means of rehabilitation for stroke survivors, and effective recognition of human motion intentions is a key premise in controlling wearable robots. In this paper, we propose an inertial measurement unit (IMU)-based gait phase detection system. The system consists of two IMUs that are tied on the thigh and on the shank, respectively, for collecting acceleration and angular velocity. Features were extracted using a sliding window of 150 ms in length, which was then fed into a quadratic discriminant analysis (QDA) classifier for classification. We recruited five stroke survivors to test our system. They walked at their own preferred speed on the level ground. Experimental results show that our proposed system has the ability of recognizing the gait phase of stroke survivors. All recognition accuracy results are above 96.5%, and detections are about 5–15 ms in advance of time. In addition, using only one IMU can also give reliable recognition results. This paper proposes an idea about the further research on human–computer interaction for the control of wearable robots.

Download Full-text

Fall inducing movable platform (FIMP) for overground trips and slips

Journal of NeuroEngineering and Rehabilitation ◽

10.1186/s12984-020-00785-0 ◽

2020 ◽

Vol 17 (1) ◽

Author(s):

Jie Kai Er ◽

Cyril John William Donnelly ◽

Seng Kwee Wee ◽

Wei Tech Ang

Keyword(s):

Real World ◽

Muscle Activation ◽

Statistical Parametric Mapping ◽

Detection Algorithm ◽

Repeated Measure ◽

Future Research ◽

Phase Detection ◽

In Series ◽

Gait Phase ◽

Movable Platform

Abstract Background The study of falls and fall prevention/intervention devices requires the recording of true falls incidence. However, true falls are rare, random, and difficult to collect in real world settings. A system capable of producing falls in an ecologically valid manner will be very helpful in collecting the data necessary to advance our understanding of the neuro and musculoskeletal mechanisms underpinning real-world falls events. Methods A fall inducing movable platform (FIMP) was designed to arrest or accelerate a subject’s ankle to induce a trip or slip. The ankle was arrested posteriorly with an electromagnetic brake and accelerated anteriorly with a motor. A power spring was connected in series between the ankle and the brake/motor to allow freedom of movement (system transparency) when a fall is not being induced. A gait phase detection algorithm was also created to enable precise activation of the fall inducing mechanisms. Statistical Parametric Mapping (SPM1D) and one-way repeated measure ANOVA were used to evaluate the ability of the FIMP to induce a trip or slip. Results During FIMP induced trips, the brake activates at the terminal swing or mid swing gait phase to induce the lowering or skipping strategies, respectively. For the lowering strategy, the characteristic leg lowering and subsequent contralateral leg swing was seen in all subjects. Likewise, for the skipping strategy, all subjects skipped forward on the perturbed leg. Slip was induced by FIMP by using a motor to impart unwanted forward acceleration to the ankle with the help of friction-reducing ground sliding sheets. Joint stiffening was observed during the slips, and subjects universally adopted the surfing strategy after the initial slip. Conclusion The results indicate that FIMP can induce ecologically valid falls under controlled laboratory conditions. The use of SPM1D in conjunction with FIMP allows for the time varying statistical quantification of trip and slip reactive kinematics events. With future research, fall recovery anomalies in subjects can now also be systematically evaluated through the assessment of other neuromuscular variables such as joint forces, muscle activation and muscle forces.

Download Full-text