Qin Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Qin Hu has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Seyed Farokh Atashzar
Department of Electrical and Computer Engineering, New York University (NYU), USA
6 shared publications - Golara Ahmadi Azar
Cerebras, Santa Clara, CA, USA
2 shared publications - Paras Gulati
Department of Electrical and Computer Engineering, New York University (NYU), New York, NY, USA
2 shared publications - Sundeep Rangan
NYU Tandon School of Engineering
2 shared publications - Alireza Sarmadi
Department of Electrical and Computer Engineering, New York University, New York, NY, USA
1 shared publication Alyson Fletcher
Department of Electrical and Computer Engineering and Department of Statistics, Mathematics, and Computer Science, University of California at Los Angeles (UCLA), Los Angeles, CA, USA
1 shared publication
Latest Publications
- 17 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (3) , 419-430
ViT-MDHGR: Cross-Day Reliability and Agility in Dynamic Hand Gesture Prediction via HD-sEMG Signal Decoding
Surface electromyography (sEMG) and high-density sEMG (HD-sEMG) biosignals have been extensively investigated for myoelectric control of prosthetic devices, neurorobotics, and more recently human-computer interfaces because of their capability for hand gesture recognition/prediction in a wearable and non-invasive manner. High intraday (same-day) performance has been reported. However, the interday performance (separating training and testing days) is substantially degraded due to the poor generalizability of conventional approaches over time, hindering the application of such techniques in real-life practices. There are limited recent studies on the feasibility of multi-day hand gesture recognition. The existing studies face a major challenge: the need for long sEMG epochs makes the corresponding neural interfaces impractical due to the induced delay in myoelectric control. This paper proposes a compact ViT-based network for multi-day dynamic hand gesture prediction. We tackle the main challenge as the proposed model only relies on very short HD-sEMG signal windows (i.e., 50 ms, accounting for only one-sixth of the convention for real-time myoelectric implementation), boosting agility and responsiveness. Our proposed model can predict 11 dynamic gestures for 20 subjects with an average accuracy of over 71% on the testing day, 3-25 days after training. Moreover, when calibrated on just a small portion of data from the testing day, the proposed model can achieve over 92% accuracy by retraining less than 10% of the parameters for computational efficiency.
- 17 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (3) , 419-430
ViT-MDHGR: Cross-Day Reliability and Agility in Dynamic Hand Gesture Prediction via HD-sEMG Signal Decoding
Surface electromyography (sEMG) and high-density sEMG (HD-sEMG) biosignals have been extensively investigated for myoelectric control of prosthetic devices, neurorobotics, and more recently human-computer interfaces because of their capability for hand gesture recognition/prediction in a wearable and non-invasive manner. High intraday (same-day) performance has been reported. However, the interday performance (separating training and testing days) is substantially degraded due to the poor generalizability of conventional approaches over time, hindering the application of such techniques in real-life practices. There are limited recent studies on the feasibility of multi-day hand gesture recognition. The existing studies face a major challenge: the need for long sEMG epochs makes the corresponding neural interfaces impractical due to the induced delay in myoelectric control. This paper proposes a compact ViT-based network for multi-day dynamic hand gesture prediction. We tackle the main challenge as the proposed model only relies on very short HD-sEMG signal windows (i.e., 50 ms, accounting for only one-sixth of the convention for real-time myoelectric implementation), boosting agility and responsiveness. Our proposed model can predict 11 dynamic gestures for 20 subjects with an average accuracy of over 71% on the testing day, 3-25 days after training. Moreover, when calibrated on just a small portion of data from the testing day, the proposed model can achieve over 92% accuracy by retraining less than 10% of the parameters for computational efficiency.
- 5 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
X-MyoNET: Biometric Identification Using Deep Processing of Dynamic Surface Electromyography
This paper investigates the potential of surface electromyography (sEMG) as a new biometric modality and proposes a deep neural network architecture as the backbone of a gesture-independent personal identification system. The paper focuses on the real-world translation of such a model through systematic optimization, which finds the minimum number of gestures and sensors needed for training. Focusing on “dynamic sEMG”, our proposed method can successfully identify 40 subjects with an average accuracy of 97%. This is achieved when gestures are the same in training, validation, and testing (the subjects need to repeat a particular gesture among a set of seven known gestures as a passcode). In a more complex scenario, when training gestures differ from those in validation and testing, our model can achieve an average accuracy of 90%, demonstrating that the proposed model can extract the unique patterns to identify a user regardless of gestures. Taking advantage of Gradient-weighted Class Activation Mapping (Grad-CAM), we explore the attention of the model on segments of the spectrotemporal space of the input signals. Grad-CAM not only sheds light on sEMG-based personal identification by decoding and visualizing the unique user-specific neurophysiological pattern but also generates a two-dimensional spectrotemporal mask used to reduce the model complexity significantly. As a result of the systematic optimization and Grad-CAM analysis, our proposed identification method needs only 4% of data for training, boosting practicality. This paper also reveals the robustness of the proposed model for cross-day evaluation. Finally, the comparative study shows the superiority of our proposed model over several state-of-the-art algorithms.
- 5 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
X-MyoNET: Biometric Identification Using Deep Processing of Dynamic Surface Electromyography
This paper investigates the potential of surface electromyography (sEMG) as a new biometric modality and proposes a deep neural network architecture as the backbone of a gesture-independent personal identification system. The paper focuses on the real-world translation of such a model through systematic optimization, which finds the minimum number of gestures and sensors needed for training. Focusing on “dynamic sEMG”, our proposed method can successfully identify 40 subjects with an average accuracy of 97%. This is achieved when gestures are the same in training, validation, and testing (the subjects need to repeat a particular gesture among a set of seven known gestures as a passcode). In a more complex scenario, when training gestures differ from those in validation and testing, our model can achieve an average accuracy of 90%, demonstrating that the proposed model can extract the unique patterns to identify a user regardless of gestures. Taking advantage of Gradient-weighted Class Activation Mapping (Grad-CAM), we explore the attention of the model on segments of the spectrotemporal space of the input signals. Grad-CAM not only sheds light on sEMG-based personal identification by decoding and visualizing the unique user-specific neurophysiological pattern but also generates a two-dimensional spectrotemporal mask used to reduce the model complexity significantly. As a result of the systematic optimization and Grad-CAM analysis, our proposed identification method needs only 4% of data for training, boosting practicality. This paper also reveals the robustness of the proposed model for cross-day evaluation. Finally, the comparative study shows the superiority of our proposed model over several state-of-the-art algorithms.
- 20 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (9) , 14778-14791
A Deep Learning Sequential Decoder for Transient High-Density Electromyography in Hand Gesture Recognition Using Subject-Embedded Transfer Learning
Hand gesture recognition (HGR) has gained significant attention due to the increasing use of AI-powered human-computer interfaces that can interpret the deep spatiotemporal dynamics of biosignals from the peripheral nervous system, such as surface electromyography (sEMG). These interfaces have a range of applications, including the control of extended reality, agile prosthetics, and exoskeletons. However, the natural variability of sEMG among individuals has led researchers to focus on subject-specific solutions. Deep learning methods, which often have complex structures, are particularly data-hungry and can be time-consuming to train, making them less practical for subject-specific applications. The main contribution of this paper is to propose and develop a generalizable, sequential decoder of transient high-density sEMG (HD-sEMG) that achieves 73% average accuracy on 65 gestures for partially-observed subjects through subject-embedded transfer learning, leveraging pre-knowledge of HGR acquired during pre-training. The use of transient HD-sEMG before gesture stabilization allows us to predict gestures with the ultimate goal of counterbalancing system control delays. The results show that the proposed generalized models significantly outperform subject-specific approaches, especially when the training data is limited and there is a significant number of gesture classes. By building on pre-knowledge and incorporating a multiplicative subject-embedded structure, our method comparatively achieves more than 13% average accuracy across partially-observed subjects with minimal data availability. This work highlights the potential of HD-sEMG and demonstrates the benefits of modeling common patterns across users to reduce the need for large amounts of data for new users, enhancing practicality.
- 20 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (9) , 14778-14791
A Deep Learning Sequential Decoder for Transient High-Density Electromyography in Hand Gesture Recognition Using Subject-Embedded Transfer Learning
Hand gesture recognition (HGR) has gained significant attention due to the increasing use of AI-powered human-computer interfaces that can interpret the deep spatiotemporal dynamics of biosignals from the peripheral nervous system, such as surface electromyography (sEMG). These interfaces have a range of applications, including the control of extended reality, agile prosthetics, and exoskeletons. However, the natural variability of sEMG among individuals has led researchers to focus on subject-specific solutions. Deep learning methods, which often have complex structures, are particularly data-hungry and can be time-consuming to train, making them less practical for subject-specific applications. The main contribution of this paper is to propose and develop a generalizable, sequential decoder of transient high-density sEMG (HD-sEMG) that achieves 73% average accuracy on 65 gestures for partially-observed subjects through subject-embedded transfer learning, leveraging pre-knowledge of HGR acquired during pre-training. The use of transient HD-sEMG before gesture stabilization allows us to predict gestures with the ultimate goal of counterbalancing system control delays. The results show that the proposed generalized models significantly outperform subject-specific approaches, especially when the training data is limited and there is a significant number of gesture classes. By building on pre-knowledge and incorporating a multiplicative subject-embedded structure, our method comparatively achieves more than 13% average accuracy across partially-observed subjects with minimal data availability. This work highlights the potential of HD-sEMG and demonstrates the benefits of modeling common patterns across users to reduce the need for large amounts of data for new users, enhancing practicality.
- 15 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (4) , 9216-9223
Hand Gesture Recognition via Transient sEMG Using Transfer Learning of Dilated Efficient CapsNet: Towards Generalization for Neurorobotics
There has been an accelerated surge in utilizing the deep neural network to decode central and peripheral activations of the human nervous system to boost the spatiotemporal resolution of neural interfaces used in human-centered robotic systems, such as prosthetics, and exoskeletons. Deep learning methods are proven to achieve high accuracy but are also challenged by their assumption of having access to massive training samples. Objective: In this letter, we propose Dilated Efficient CapsNet to improve the predictive performance when the available individual data is minimal and not enough to train an individualized network for controlling a personalized robotic system. Method: We proposed the concept of transfer learning for a new design of the dilated efficient capsular neural network to relax the need of having access to massive individual data and utilize the field knowledge which can be learned from a group of participants. In addition, instead of using complete sEMG signals, we only use the transient phase, reducing the volume of training samples to 20% of the original and maximizing the agility. Results: In experiments, we validate the performance with various amounts of injected personalized training data (from 25% to 100% of transient phase). The results support the use of the proposed transfer learning approach based on the dilated capsular neural network when the knowledge domain learned on a small number of subjects can be utilized to minimize the need for new data from new subjects. The model focuses only on the transient phase which is a challenging neural interfacing problem.
- 15 July 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (4) , 9216-9223
Hand Gesture Recognition via Transient sEMG Using Transfer Learning of Dilated Efficient CapsNet: Towards Generalization for Neurorobotics
There has been an accelerated surge in utilizing the deep neural network to decode central and peripheral activations of the human nervous system to boost the spatiotemporal resolution of neural interfaces used in human-centered robotic systems, such as prosthetics, and exoskeletons. Deep learning methods are proven to achieve high accuracy but are also challenged by their assumption of having access to massive training samples. Objective: In this letter, we propose Dilated Efficient CapsNet to improve the predictive performance when the available individual data is minimal and not enough to train an individualized network for controlling a personalized robotic system. Method: We proposed the concept of transfer learning for a new design of the dilated efficient capsular neural network to relax the need of having access to massive individual data and utilize the field knowledge which can be learned from a group of participants. In addition, instead of using complete sEMG signals, we only use the transient phase, reducing the volume of training samples to 20% of the original and maximizing the agility. Results: In experiments, we validate the performance with various amounts of injected personalized training data (from 25% to 100% of transient phase). The results support the use of the proposed transfer learning approach based on the dilated capsular neural network when the knowledge domain learned on a small number of subjects can be utilized to minimize the need for new data from new subjects. The model focuses only on the transient phase which is a challenging neural interfacing problem.
- 13 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 2851-2858
Deep Heterogeneous Dilation of LSTM for Transient-Phase Gesture Prediction Through High-Density Electromyography: Towards Application in Neurorobotics
Deep networks have been recently proposed to estimate motor intention using conventional bipolar surface electromyography (sEMG) signals for myoelectric control of neurorobots. In this regard, Deepnets are generally challenged by long training times (affecting practicality and calibration), complex model architectures (affecting the predictability of the outcomes), and a large number of trainable parameters (increasing the need for Big Data). Capitalizing on our recent work on homogeneous temporal dilation in a Recurrent Neural Network (RNN) model, this letter proposes, for the first time, heterogeneous temporal dilation in an LSTM model and applies that to high-density surface electromyography (HD-sEMG), allowing for the decoding of dynamic temporal dependencies with tunable temporal foci. In this letter, a 128-channel HD-sEMG signal space is considered due to the potential for enhancing the spatiotemporal resolution of human-robot interfaces. Accordingly, this letter addresses a challenging motor intention decoding problem of neurorobots, namely, transient intention identification . Our approach uses only the dynamic and transient phase of gesture movements when the signals are not stabilized or plateaued, which can significantly enhance the temporal resolution of human-robot interfaces. This would eventually enhance seamless real-time implementations. Additionally, this letter introduces the concept of “dilation foci” to modulate the modeling of temporal variation in transient phases. In this work a high number (e.g., 65) of gestures is included, which adds to the complexity and significance of the understudied problem. Our results show state-of-the-art performance for gesture prediction in terms of accuracy, training time, and model convergence.
- 13 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 7 (2) , 2851-2858
Deep Heterogeneous Dilation of LSTM for Transient-Phase Gesture Prediction Through High-Density Electromyography: Towards Application in Neurorobotics
Deep networks have been recently proposed to estimate motor intention using conventional bipolar surface electromyography (sEMG) signals for myoelectric control of neurorobots. In this regard, Deepnets are generally challenged by long training times (affecting practicality and calibration), complex model architectures (affecting the predictability of the outcomes), and a large number of trainable parameters (increasing the need for Big Data). Capitalizing on our recent work on homogeneous temporal dilation in a Recurrent Neural Network (RNN) model, this letter proposes, for the first time, heterogeneous temporal dilation in an LSTM model and applies that to high-density surface electromyography (HD-sEMG), allowing for the decoding of dynamic temporal dependencies with tunable temporal foci. In this letter, a 128-channel HD-sEMG signal space is considered due to the potential for enhancing the spatiotemporal resolution of human-robot interfaces. Accordingly, this letter addresses a challenging motor intention decoding problem of neurorobots, namely, transient intention identification . Our approach uses only the dynamic and transient phase of gesture movements when the signals are not stabilized or plateaued, which can significantly enhance the temporal resolution of human-robot interfaces. This would eventually enhance seamless real-time implementations. Additionally, this letter introduces the concept of “dilation foci” to modulate the modeling of temporal variation in transient phases. In this work a high number (e.g., 65) of gestures is included, which adds to the complexity and significance of the understudied problem. Our results show state-of-the-art performance for gesture prediction in terms of accuracy, training time, and model convergence.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (4) , 6212-6219
Temporal Dilation of Deep LSTM for Agile Decoding of sEMG: Application in Prediction of Upper-Limb Motor Intention in NeuroRobotics
The spectrotemporal information content of surface electromyography has shown strong potential in predicting the intended motor command. During the last decade, with accelerated exploitation of powerful deep-learning techniques aligned with advancements in active prostheses and neurorobots, a great deal of interest has been drawn to the development of intelligent myoelectric prostheses with an ultimate resolution of upper-limb gestures prediction. Recent research involves Deep CNNs, RNNs, and hybrid frameworks, which have shown promising results. However, deep-learning models have almost always been challenged by the structural complexity, the large number of trainable parameters, concerns of overfitting, and prolonged training time, which complicate the practicality and limit the outcomes. In this letter, for the first time, we propose temporal-dilation in the LSTM module of a hybrid Deepnet model for sEMG-based gesture detection, hypothesizing improved accuracy and training agility. We also analyze the effect of dilation-aggressiveness. We conduct systematic and statistical analysis on the efficacy of the proposed approach in comparison to recent literature, including our previous work. this letter shows that the proposed temporally-dilated LSTM model wins over the recent deep-learning techniques in terms of accuracy, and more significantly, it reduces the training time while increasing the convergence speed, with the ultimate goal of maximizing practicality and translational value for neurorobotic systems.
- 23 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (4) , 6212-6219
Temporal Dilation of Deep LSTM for Agile Decoding of sEMG: Application in Prediction of Upper-Limb Motor Intention in NeuroRobotics
The spectrotemporal information content of surface electromyography has shown strong potential in predicting the intended motor command. During the last decade, with accelerated exploitation of powerful deep-learning techniques aligned with advancements in active prostheses and neurorobots, a great deal of interest has been drawn to the development of intelligent myoelectric prostheses with an ultimate resolution of upper-limb gestures prediction. Recent research involves Deep CNNs, RNNs, and hybrid frameworks, which have shown promising results. However, deep-learning models have almost always been challenged by the structural complexity, the large number of trainable parameters, concerns of overfitting, and prolonged training time, which complicate the practicality and limit the outcomes. In this letter, for the first time, we propose temporal-dilation in the LSTM module of a hybrid Deepnet model for sEMG-based gesture detection, hypothesizing improved accuracy and training agility. We also analyze the effect of dilation-aggressiveness. We conduct systematic and statistical analysis on the efficacy of the proposed approach in comparison to recent literature, including our previous work. this letter shows that the proposed temporally-dilated LSTM model wins over the recent deep-learning techniques in terms of accuracy, and more significantly, it reduces the training time while increasing the convergence speed, with the ultimate goal of maximizing practicality and translational value for neurorobotic systems.