Junyan Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Junyan Li has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Calvin Yu-Chian Chen
AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China
2 shared publications Hsin-Yi Chen
School of Intelligent Systems Engineering, Artificial Intelligence Medical Center , Sun Yat-sen University , Shenzhen 510275 , China
2 shared publicationsBingxuan Qiao
Key Laboratory for Corrosion and Protection (MOE, School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing, China
1 shared publicationChaofang Dong
Key Laboratory for Corrosion and Protection (MOE, School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing, China
1 shared publication- Chengpei Tang
School of Intelligent Systems Engineering, Sun Yat-sen University, Guangzhou, China
1 shared publication - Chun Zhao
School of Advanced Technology Xi'an Jiaotong‐Liverpool University Suzhou China
1 shared publication
Latest Publications
- 28 February 2025
- Published by Wiley in Advanced Optical Materials
- Vol. 13 (13)
Quantum Dot‐Enhanced Dual‐Modality Heterojunction Optoelectronic Synapse for Neuromorphic Computing
The advancement of optoelectronic sensing synapse devices, which integrate multiple sensory modalities and achieve the efficiency of biological vision systems, is crucial for the field of artificial vision systems. This work incorporates CdSe/CdSe x S 1‐x quantum dots with In 2 O 3 semiconductor into a heterojunction via low‐cost fully solution‐based process to endow the synaptic transistor with dual‐modality of lights and electricity. Optoelectronic synaptic transistors exhibit sensitivity to a broad spectrum of light, encompassing wavelengths ranging from 395 to 808 nm, in addition to their responsiveness to electrical signals. The efficiency of information processing is therefore improved by integration of senses. Additionally, by doping lithium ions into the dielectric layer, the gate capacitance is increased by over ten times and significantly improved the devices channel modulation and retention characteristics. An artificial visual perception demo based on the Quantum Dot‐Enhanced synaptic ransistors (QDET) is well presented to showcase their practical application in pattern recognition and QDETs offer a promising platform for energy‐efficient, high‐performance neuromorphic systems.
- 28 February 2025
- Published by Wiley in Advanced Optical Materials
- Vol. 13 (13)
Quantum Dot‐Enhanced Dual‐Modality Heterojunction Optoelectronic Synapse for Neuromorphic Computing
The advancement of optoelectronic sensing synapse devices, which integrate multiple sensory modalities and achieve the efficiency of biological vision systems, is crucial for the field of artificial vision systems. This work incorporates CdSe/CdSe x S 1‐x quantum dots with In 2 O 3 semiconductor into a heterojunction via low‐cost fully solution‐based process to endow the synaptic transistor with dual‐modality of lights and electricity. Optoelectronic synaptic transistors exhibit sensitivity to a broad spectrum of light, encompassing wavelengths ranging from 395 to 808 nm, in addition to their responsiveness to electrical signals. The efficiency of information processing is therefore improved by integration of senses. Additionally, by doping lithium ions into the dielectric layer, the gate capacitance is increased by over ten times and significantly improved the devices channel modulation and retention characteristics. An artificial visual perception demo based on the Quantum Dot‐Enhanced synaptic ransistors (QDET) is well presented to showcase their practical application in pattern recognition and QDETs offer a promising platform for energy‐efficient, high‐performance neuromorphic systems.
- Open Access
- 1 July 2021
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1978 (1) , 012039
Dynamic bifurcation for a three-species cooperating model
A dynamic bifurcation analysis on a three-species cooperating model was presented and it was proved that the problem bifurcated an attractor as the parameter λ crossed the critical value λ0. The analysis was based on the attractor bifurcation theory together with the central manifold reduction.
- 1 July 2021
- Published by IOP Publishing in Journal of Physics: Conference Series
- Vol. 1978 (1) , 012039
Open AccessDynamic bifurcation for a three-species cooperating model
A dynamic bifurcation analysis on a three-species cooperating model was presented and it was proved that the problem bifurcated an attractor as the parameter λ crossed the critical value λ0. The analysis was based on the attractor bifurcation theory together with the central manifold reduction.
- Open Access
SlideAugment: A Simple Data Processing Method to Enhance Human Activity Recognition Accuracy Based on WiFi
Currently, there are various works presented in the literature regarding the activity recognition based on WiFi. We observe that existing public data sets do not have enough data. In this work, we present a data augmentation method called window slicing. By slicing the original data, we get multiple samples for one raw datum. As a result, the size of the data set can be increased. On the basis of the experiments performed on a public data set and our collected data set, we observe that the proposed method assists in improving the results. It is notable that, on the public data set, the activity recognition accuracy improves from 88.13% to 97.12%. Similarly, the recognition accuracy is also improved for the data set collected in this work. Although the proposed method is simple, it effectively enhances the recognition accuracy. It is a general channel state information (CSI) data augmentation method. In addition, the proposed method demonstrates good interpretability.
Open AccessSlideAugment: A Simple Data Processing Method to Enhance Human Activity Recognition Accuracy Based on WiFi
Currently, there are various works presented in the literature regarding the activity recognition based on WiFi. We observe that existing public data sets do not have enough data. In this work, we present a data augmentation method called window slicing. By slicing the original data, we get multiple samples for one raw datum. As a result, the size of the data set can be increased. On the basis of the experiments performed on a public data set and our collected data set, we observe that the proposed method assists in improving the results. It is notable that, on the public data set, the activity recognition accuracy improves from 88.13% to 97.12%. Similarly, the recognition accuracy is also improved for the data set collected in this work. Although the proposed method is simple, it effectively enhances the recognition accuracy. It is a general channel state information (CSI) data augmentation method. In addition, the proposed method demonstrates good interpretability.
- 30 November 2020
- Published by Emerald Publishing in Anti-Corrosion Methods and Materials
- Vol. 67 (6) , 575-581
The mechanical properties and corrosion resistance of selective laser melting 30CrMnSiA steel
Purpose: The purpose of this paper is to study the mechanical and corrosion behaviors of selective laser melting (SLM) 30CrMnSiA steel.Design/methodology/approach: The microstructure, mechanical behavior and corrosion resistance of SLM 30CrMnSiA steel were studied by electrochemical test and immersion experiment.Findings: The results showed that the microstructure of SLM 30CrMnSiA steel contained equiaxed fine (25.0 µm) ferrite and lamellar pearlite. The tensile strength of SLM 30CrMnSiA steel under the process parameters is 1,447 MPa and the elongation is 7.5%. As the immersion/salt spray test time increased to 15 days/48 h, corrosion mainly occurred in the local position of the sample and corrosion resistance decreased. When the immersion/salt spray test time increased to 30 days/168 h, the corrosion production covered the surface of the samples and the corrosion resistance increased.Originality/value: The paper can guide the application andin siturepair for low-alloy steel by additive manufacturing.
- 30 November 2020
- Published by Emerald Publishing in Anti-Corrosion Methods and Materials
- Vol. 67 (6) , 575-581
The mechanical properties and corrosion resistance of selective laser melting 30CrMnSiA steel
Purpose: The purpose of this paper is to study the mechanical and corrosion behaviors of selective laser melting (SLM) 30CrMnSiA steel.Design/methodology/approach: The microstructure, mechanical behavior and corrosion resistance of SLM 30CrMnSiA steel were studied by electrochemical test and immersion experiment.Findings: The results showed that the microstructure of SLM 30CrMnSiA steel contained equiaxed fine (25.0 µm) ferrite and lamellar pearlite. The tensile strength of SLM 30CrMnSiA steel under the process parameters is 1,447 MPa and the elongation is 7.5%. As the immersion/salt spray test time increased to 15 days/48 h, corrosion mainly occurred in the local position of the sample and corrosion resistance decreased. When the immersion/salt spray test time increased to 30 days/168 h, the corrosion production covered the surface of the samples and the corrosion resistance increased.Originality/value: The paper can guide the application andin siturepair for low-alloy steel by additive manufacturing.
- 14 August 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 10 (17) , 4947-4961
Artificial Intelligence Approach To Investigate the Longevity Drug
Longevity is a very important and interesting topic, and Klotho has been demonstrated to be related to longevity. We combined network pharmacology, machine learning, deep learning, and molecular dynamics (MD) simulation to investigate potent lead drugs. Related protein insulin-like growth factor 1 receptor (IGF1R) and insulin receptor (IR) were docked with the traditional Chinese medicine (TCM) database to screen out several novel candidates. Besides, nine different machine learning algorithms were performed to build reliable and accurate predicted models. Moreover, we used the novel deep learning algorithm to build predicted models. All of these models obtained significant R2, which are all greater than 0.87 on the training set and higher than 0.88 for the test set, respectively. The long time 500 ns molecular dynamics simulation was also performed to verify protein–ligand properties and stability. Finally, we obtained Antifebrile Dichroa, Holarrhena antidysenterica, and Gelsemium sempervirens, which might be potent TCMs for two targets.
- 14 August 2019
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry Letters
- Vol. 10 (17) , 4947-4961
Artificial Intelligence Approach To Investigate the Longevity Drug
Longevity is a very important and interesting topic, and Klotho has been demonstrated to be related to longevity. We combined network pharmacology, machine learning, deep learning, and molecular dynamics (MD) simulation to investigate potent lead drugs. Related protein insulin-like growth factor 1 receptor (IGF1R) and insulin receptor (IR) were docked with the traditional Chinese medicine (TCM) database to screen out several novel candidates. Besides, nine different machine learning algorithms were performed to build reliable and accurate predicted models. Moreover, we used the novel deep learning algorithm to build predicted models. All of these models obtained significant R2, which are all greater than 0.87 on the training set and higher than 0.88 for the test set, respectively. The long time 500 ns molecular dynamics simulation was also performed to verify protein–ligand properties and stability. Finally, we obtained Antifebrile Dichroa, Holarrhena antidysenterica, and Gelsemium sempervirens, which might be potent TCMs for two targets.
- 19 March 2019
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 59 (4) , 1605-1623
Deep Learning and Random Forest Approach for Finding the Optimal Traditional Chinese Medicine Formula for Treatment of Alzheimer’s Disease
It has demonstrated that Glycogen synthase kinase-3β (GSK3β) is related with Alzheimer’s disease (AD). Based on the world largest traditional Chinese medicine (TCM) database, network pharmacology-based approach were utilized to investigate the TCM candidates that can dock well in multiple targets. Support Vector Machine(SVM), multiple linear regression (MLR) methods were utilized to obtain predicted models, Specially, The Deep Learning and The Random Forest algorithm were adopted, We achieved R2 of 0.927 on training set and 0.862 on the test set in Deep Learning, The latter’s R2 on the training set is 0.869 and on the test set is 0.890.Besides, comparative molecular similarity indices analysis (CoMSIA) was performed to get a predicted model. All the training models achieved good results on test set.100ns MD simulation evaluated the stability of GSK3β protein-ligand complexes. Methyl 3-O-feruloylquinate and Cynanogenin A both induced more compactness to GSK3β complex and stable condition among all simulation times, the GSK3β complex also had no substantial fluctuation after simulation time of 5 ns. For TCM molecules, we used trained models to calculate predicted bioactivity values.Thus, the optimum TCM candidates were obtained by ranking the predicted values.The result showed Methyl 3-O-feruloylquinate contained in Phellodendron amurense and Cynanogenin A contained in Cynanchum atratum are capable of forming stable interactions with GSK3β.
- 19 March 2019
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 59 (4) , 1605-1623
Deep Learning and Random Forest Approach for Finding the Optimal Traditional Chinese Medicine Formula for Treatment of Alzheimer’s Disease
It has demonstrated that Glycogen synthase kinase-3β (GSK3β) is related with Alzheimer’s disease (AD). Based on the world largest traditional Chinese medicine (TCM) database, network pharmacology-based approach were utilized to investigate the TCM candidates that can dock well in multiple targets. Support Vector Machine(SVM), multiple linear regression (MLR) methods were utilized to obtain predicted models, Specially, The Deep Learning and The Random Forest algorithm were adopted, We achieved R2 of 0.927 on training set and 0.862 on the test set in Deep Learning, The latter’s R2 on the training set is 0.869 and on the test set is 0.890.Besides, comparative molecular similarity indices analysis (CoMSIA) was performed to get a predicted model. All the training models achieved good results on test set.100ns MD simulation evaluated the stability of GSK3β protein-ligand complexes. Methyl 3-O-feruloylquinate and Cynanogenin A both induced more compactness to GSK3β complex and stable condition among all simulation times, the GSK3β complex also had no substantial fluctuation after simulation time of 5 ns. For TCM molecules, we used trained models to calculate predicted bioactivity values.Thus, the optimum TCM candidates were obtained by ranking the predicted values.The result showed Methyl 3-O-feruloylquinate contained in Phellodendron amurense and Cynanogenin A contained in Cynanchum atratum are capable of forming stable interactions with GSK3β.