Jianhua Wang
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
Research Subjects
Jianhua Wang has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Xiaolin Chang
Beijing Jiaotong University
6 shared publications - Jelena Mišić
Department of Computer Science, Toronto Metropolitan University, Toronto, ON, Canada
4 shared publications - Vojislav B. Misic
Department of Computer Science, Toronto Metropolitan University, Toronto, ON, Canada
4 shared publications - Junchao Fan
Beijing Jiaotong University
2 shared publications - Lin Li
Beijing Jiaotong University
2 shared publications Zhihui Zhao
2 shared publications
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (18) , 43540-43553
Sparse Threats, Focused Defense: Criticality-Aware Robust Reinforcement Learning for Safe Autonomous Driving
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (18) , 43540-43553
Sparse Threats, Focused Defense: Criticality-Aware Robust Reinforcement Learning for Safe Autonomous Driving
- Open Access
- 13 July 2026
- Published by Springer Nature in Cybersecurity
- Vol. 9 (1)
Feddsg: backdoor defense via semantic filter and geometric constraint in federated learning
Backdoor attacks pose a serious threat to Internet-of-Things (IoT) federated learning. In IoT deployments, pronounced non-independent and identically distributed (non-IID) data heterogeneity causes benign client updates to exhibit substantial variability across devices. Meanwhile, the physical exposure of IoT devices increases the risk of large-scale compromise and elevated malicious participation. Such variability allows poisoned updates to blend into natural fluctuations, rendering many robust aggregation and detection-based defenses unreliable. We propose FedDSG , a server-side defense that combines a semantic bias filter and a geometric direction constraint to counter backdoor manipulation. FedDSG first extracts a novel scale-invariant semantic cue from the last-layer bias of client updates to identify abnormal target-class reinforcement, staying effective even when benign bias patterns differ substantially across clients. The remaining updates are then constrained using a reference derived from a small trusted anchor set, limiting adversarial drift. This sequential design links semantic cues with geometric structure, where the former removes clearly suspicious updates and the latter stabilizes the residual ones, preventing misdetection-induced drift amplification while avoiding distortion of benign updates. The method does not alter client behavior or communication and adds minimal server-side overhead. Extensive experiments on MNIST, Fashion-MNIST, CIFAR-10, and SVHN under non-IID distributions with high malicious participation demonstrate the robustness of FedDSG. It reduces the attack success rate to 0.003, 0.006, 0.007, and 0.091, respectively, with only marginal accuracy loss and consistently achieves the highest Overall Performance Score (OPS), reflecting a superior trade-off between robustness and accuracy. Code and data availability information is provided in the Availability of data and materials section.
- 13 July 2026
- Published by Springer Nature in Cybersecurity
- Vol. 9 (1)
Open AccessFeddsg: backdoor defense via semantic filter and geometric constraint in federated learning
Backdoor attacks pose a serious threat to Internet-of-Things (IoT) federated learning. In IoT deployments, pronounced non-independent and identically distributed (non-IID) data heterogeneity causes benign client updates to exhibit substantial variability across devices. Meanwhile, the physical exposure of IoT devices increases the risk of large-scale compromise and elevated malicious participation. Such variability allows poisoned updates to blend into natural fluctuations, rendering many robust aggregation and detection-based defenses unreliable. We propose FedDSG , a server-side defense that combines a semantic bias filter and a geometric direction constraint to counter backdoor manipulation. FedDSG first extracts a novel scale-invariant semantic cue from the last-layer bias of client updates to identify abnormal target-class reinforcement, staying effective even when benign bias patterns differ substantially across clients. The remaining updates are then constrained using a reference derived from a small trusted anchor set, limiting adversarial drift. This sequential design links semantic cues with geometric structure, where the former removes clearly suspicious updates and the latter stabilizes the residual ones, preventing misdetection-induced drift amplification while avoiding distortion of benign updates. The method does not alter client behavior or communication and adds minimal server-side overhead. Extensive experiments on MNIST, Fashion-MNIST, CIFAR-10, and SVHN under non-IID distributions with high malicious participation demonstrate the robustness of FedDSG. It reduces the attack success rate to 0.003, 0.006, 0.007, and 0.091, respectively, with only marginal accuracy loss and consistently achieves the highest Overall Performance Score (OPS), reflecting a superior trade-off between robustness and accuracy. Code and data availability information is provided in the Availability of data and materials section.
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (7) , 10590-10606
SMTFL: Secure Model Training to Untrusted Participants in Federated Learning
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (7) , 10590-10606
SMTFL: Secure Model Training to Untrusted Participants in Federated Learning
- 24 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 22 (1) , 198-208
CRS-FL: Conditional Random Sampling for Communication-Efficient and Privacy-Preserving Federated Learning
Federated Learning (FL), a privacy-oriented distributed ML paradigm, is gaining great interest in the Internet of Things because of its capability to protect participants’ data privacy. Studies have been conducted to address the challenges of communication efficiency and privacy-preserving, which exist in standard FL. However, they cannot achieve the goal of making a tradeoff between communication efficiency and model accuracy while guaranteeing privacy. This paper proposes a Conditional Random Sampling (CRS) method and implements it into the standard FL (CRS-FL) to tackle the above-mentioned challenges. CRS explores a Poisson-sampling-based stochastic coefficient to achieve a higher probability of obtaining zero-gradient unbiasedly and then decreases the communication overhead effectively without model accuracy degradation. Moreover, we dig out the relaxation Local Differential Privacy (LDP) guarantee conditions of CRS theoretically. Extensive experiment results indicate that (1) in communication efficiency, CRS-FL performs better than the existing methods in metric accuracy per transmission byte without model accuracy reduction in more than 7% sampling ratio (# sampling size / # model size); (2) in privacy-preserving, CRS-FL achieves no accuracy reduction compared with LDP baselines while holding the efficiency, even exceeding them in model accuracy under more sampling ratio conditions.
- 24 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 22 (1) , 198-208
CRS-FL: Conditional Random Sampling for Communication-Efficient and Privacy-Preserving Federated Learning
Federated Learning (FL), a privacy-oriented distributed ML paradigm, is gaining great interest in the Internet of Things because of its capability to protect participants’ data privacy. Studies have been conducted to address the challenges of communication efficiency and privacy-preserving, which exist in standard FL. However, they cannot achieve the goal of making a tradeoff between communication efficiency and model accuracy while guaranteeing privacy. This paper proposes a Conditional Random Sampling (CRS) method and implements it into the standard FL (CRS-FL) to tackle the above-mentioned challenges. CRS explores a Poisson-sampling-based stochastic coefficient to achieve a higher probability of obtaining zero-gradient unbiasedly and then decreases the communication overhead effectively without model accuracy degradation. Moreover, we dig out the relaxation Local Differential Privacy (LDP) guarantee conditions of CRS theoretically. Extensive experiment results indicate that (1) in communication efficiency, CRS-FL performs better than the existing methods in metric accuracy per transmission byte without model accuracy reduction in more than 7% sampling ratio (# sampling size / # model size); (2) in privacy-preserving, CRS-FL achieves no accuracy reduction compared with LDP baselines while holding the efficiency, even exceeding them in model accuracy under more sampling ratio conditions.
- 30 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 17960-17974
PA-iMFL: Communication-Efficient Privacy Amplification Method Against Data Reconstruction Attack in Improved Multilayer Federated Learning
Recently, big data has seen explosive growth in the Internet of Things (IoT). Multi-layer FL (MFL) based on cloud-edge-end architecture can promote model training efficiency and model accuracy while preserving IoT data privacy. This paper considers an improved MFL, where edge layer devices own private data and can join the training process. iMFL can improve edge resource utilization and also alleviate the strict requirement of end devices, but suffers from the issues of Data Reconstruction Attack (DRA) and unacceptable communication overhead. This paper aims to address these issues with iMFL. We propose a Privacy Amplification scheme on iMFL (PA-iMFL). Differing from standard MFL, we design privacy operations in end and edge devices after local training, including three sequential components, local differential privacy with Laplace mechanism, privacy amplification subsample, and gradient sign reset. Benefitting from privacy operations, PA-iMFL reduces communication overhead and achieves privacy-preserving. Extensive results demonstrate that against State-Of-The-Art (SOTA) DRAs, PA-iMFL can effectively mitigate private data leakage and reach the same level of protection capability as the SOTA defense model. Moreover, due to adopting privacy operations in edge devices, PA-iMFL promotes up to 2.8 × communication efficiency than the SOTA compression method without compromising model accuracy.
- 30 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (10) , 17960-17974
PA-iMFL: Communication-Efficient Privacy Amplification Method Against Data Reconstruction Attack in Improved Multilayer Federated Learning
Recently, big data has seen explosive growth in the Internet of Things (IoT). Multi-layer FL (MFL) based on cloud-edge-end architecture can promote model training efficiency and model accuracy while preserving IoT data privacy. This paper considers an improved MFL, where edge layer devices own private data and can join the training process. iMFL can improve edge resource utilization and also alleviate the strict requirement of end devices, but suffers from the issues of Data Reconstruction Attack (DRA) and unacceptable communication overhead. This paper aims to address these issues with iMFL. We propose a Privacy Amplification scheme on iMFL (PA-iMFL). Differing from standard MFL, we design privacy operations in end and edge devices after local training, including three sequential components, local differential privacy with Laplace mechanism, privacy amplification subsample, and gradient sign reset. Benefitting from privacy operations, PA-iMFL reduces communication overhead and achieves privacy-preserving. Extensive results demonstrate that against State-Of-The-Art (SOTA) DRAs, PA-iMFL can effectively mitigate private data leakage and reach the same level of protection capability as the SOTA defense model. Moreover, due to adopting privacy operations in edge devices, PA-iMFL promotes up to 2.8 × communication efficiency than the SOTA compression method without compromising model accuracy.
- 22 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 73 (4) , 1138-1151
Towards Secure Runtime Customizable Trusted Execution Environment on FPGA-SoC
Processing sensitive data and deploying well-designed Intellectual Property (IP) cores on remote Field Programmable Gate Array (FPGA) are prone to private data leakage and IP theft. One effective solution is constructing Trusted Execution Environment (TEE) and its secure boot process on FPGA-SoC (FPGA System on Chip). This paper aims to establish Secure Runtime Customizable TEE (SrcTEE) on FPGA-SoC through the design of a novel secure boot scheme and the design of the following three components: 1) CrloadIP, which enforces access control on TEE applications deploying IP at runtime such that SrcTEE can alleviate threats from unauthorized TEE applications and then SrcTEE can be adjusted dynamically and securely; 2) CexecIP, which not only enables the execution of newly-installed IP cores without modifying the operating system of FPGA-SoC TEE, but also prevents insider attacks from executing IPs in SrcTEE; 3) CremoAT, which can provide the newly-measured SrcTEE state and establish a secure communication path between remote verifiers and SrcTEE. Our secure boot scheme supports refreshable root trust key, and assures the authenticity and integrity of boot codes during the SrcTEE booting process. We conduct a security analysis of SrcTEE and its performance evaluation on Xilinx Zynq UltraScale+ XCZU15EG 2FFVB1156 MPSoC.
- 22 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 73 (4) , 1138-1151
Towards Secure Runtime Customizable Trusted Execution Environment on FPGA-SoC
Processing sensitive data and deploying well-designed Intellectual Property (IP) cores on remote Field Programmable Gate Array (FPGA) are prone to private data leakage and IP theft. One effective solution is constructing Trusted Execution Environment (TEE) and its secure boot process on FPGA-SoC (FPGA System on Chip). This paper aims to establish Secure Runtime Customizable TEE (SrcTEE) on FPGA-SoC through the design of a novel secure boot scheme and the design of the following three components: 1) CrloadIP, which enforces access control on TEE applications deploying IP at runtime such that SrcTEE can alleviate threats from unauthorized TEE applications and then SrcTEE can be adjusted dynamically and securely; 2) CexecIP, which not only enables the execution of newly-installed IP cores without modifying the operating system of FPGA-SoC TEE, but also prevents insider attacks from executing IPs in SrcTEE; 3) CremoAT, which can provide the newly-measured SrcTEE state and establish a secure communication path between remote verifiers and SrcTEE. Our secure boot scheme supports refreshable root trust key, and assures the authenticity and integrity of boot codes during the SrcTEE booting process. We conduct a security analysis of SrcTEE and its performance evaluation on Xilinx Zynq UltraScale+ XCZU15EG 2FFVB1156 MPSoC.
- Open Access
Correlation Cube Attack Revisited: Improved Cube Search and Superpoly Recovery Techniques
In this paper, we improve the cube attack by exploiting low-degree factors of the superpoly w.r.t. certain "special" index set of cube (ISoC). This can be viewed as a special case of the correlation cube attack proposed at Eurocrypt 2018, but under our framework more beneficial equations on the key variables can be obtained in the key-recovery phase. To mount our attack, one has two challenging problems: effectively recover algebraic normal form of the superpoly and extract out its low-degree factors; and efficiently search a large quantity of good ISoCs. We bring in new techniques to solve both of them. First, we propose the variable substitution technique for middle rounds of a cipher, in which polynomials on the key variables in the algebraic expressions of internal states are substituted by new variables. This will improve computational complexity of the superpoly recovery and promise more compact superpolys that can be easily decomposed with respect to the new variables. Second, we propose the vector numeric mapping technique, which seeks out a tradeoff between efficiency of the numeric mapping technique and accuracy of the monomial prediction technique in degree evaluation of superpolys. Combining with this technique, a fast pruning method is given and modeled by MILP to filter good ISoCs of which the algebraic degree satisfies some fixed threshold. Thanks to automated MILP solvers, it becomes practical to comprehensively search for good cubes across the entire search space. To illustrate the power of our techniques, we apply all of them to Trivium stream cipher. The previous best practical key recovery attack was on 820-round Trivium with complexity $2^{53.17}$. We put forward 820-, 825- and 830-round practical key-recovery attacks, in which there are 2^{80}\times 87.8%, 2^{80}\times 83% and 2^{80}\times 65.7% keys that could be practically recovered, respectively.Open AccessCorrelation Cube Attack Revisited: Improved Cube Search and Superpoly Recovery Techniques
In this paper, we improve the cube attack by exploiting low-degree factors of the superpoly w.r.t. certain "special" index set of cube (ISoC). This can be viewed as a special case of the correlation cube attack proposed at Eurocrypt 2018, but under our framework more beneficial equations on the key variables can be obtained in the key-recovery phase. To mount our attack, one has two challenging problems: effectively recover algebraic normal form of the superpoly and extract out its low-degree factors; and efficiently search a large quantity of good ISoCs. We bring in new techniques to solve both of them. First, we propose the variable substitution technique for middle rounds of a cipher, in which polynomials on the key variables in the algebraic expressions of internal states are substituted by new variables. This will improve computational complexity of the superpoly recovery and promise more compact superpolys that can be easily decomposed with respect to the new variables. Second, we propose the vector numeric mapping technique, which seeks out a tradeoff between efficiency of the numeric mapping technique and accuracy of the monomial prediction technique in degree evaluation of superpolys. Combining with this technique, a fast pruning method is given and modeled by MILP to filter good ISoCs of which the algebraic degree satisfies some fixed threshold. Thanks to automated MILP solvers, it becomes practical to comprehensively search for good cubes across the entire search space. To illustrate the power of our techniques, we apply all of them to Trivium stream cipher. The previous best practical key recovery attack was on 820-round Trivium with complexity $2^{53.17}$. We put forward 820-, 825- and 830-round practical key-recovery attacks, in which there are 2^{80}\times 87.8%, 2^{80}\times 83% and 2^{80}\times 65.7% keys that could be practically recovered, respectively. - Open Access
PASS: A Parameter Audit-based Secure and Fair Federated Learning Scheme against Free-Rider Attack
Federated Learning (FL) as a secure distributed learning framework gains interests in Internet of Things (IoT) due to its capability of protecting the privacy of participant data. However, traditional FL systems are vulnerable to Free-Rider (FR) attacks, which causes unfairness, privacy leakage and inferior performance to FL systems. The prior defense mechanisms against FR attacks assumed that malicious clients (namely, adversaries) declare less than 50% of the total amount of clients. Moreover, they aimed for Anonymous FR (AFR) attacks and lost effectiveness in resisting Selfish FR (SFR) attacks. In this paper, we propose a Parameter Audit-based Secure and fair federated learning Scheme (PASS) against FR attack. PASS has the following key features: (a) prevent from privacy leakage with less accuracy loss; (b) be effective in countering both AFR and SFR attacks; (c) work well no matter whether AFR and SFR adversaries occupy the majority of clients or not. Extensive experimental results validate that PASS: (a) has the same level as the State-Of-The-Art method in mean square error against privacy leakage; (b) defends against AFR and SFR attacks in terms of a higher defense success rate, lower false positive rate, and higher F1-score; (c) is still effective where adversaries exceed 50%, with F1-score 89% against AFR attack and F1-score 87% against SFR attack. Note that PASS produces no negative effect on FL accuracy when there is no FR adversary.
Open AccessPASS: A Parameter Audit-based Secure and Fair Federated Learning Scheme against Free-Rider Attack
Federated Learning (FL) as a secure distributed learning framework gains interests in Internet of Things (IoT) due to its capability of protecting the privacy of participant data. However, traditional FL systems are vulnerable to Free-Rider (FR) attacks, which causes unfairness, privacy leakage and inferior performance to FL systems. The prior defense mechanisms against FR attacks assumed that malicious clients (namely, adversaries) declare less than 50% of the total amount of clients. Moreover, they aimed for Anonymous FR (AFR) attacks and lost effectiveness in resisting Selfish FR (SFR) attacks. In this paper, we propose a Parameter Audit-based Secure and fair federated learning Scheme (PASS) against FR attack. PASS has the following key features: (a) prevent from privacy leakage with less accuracy loss; (b) be effective in countering both AFR and SFR attacks; (c) work well no matter whether AFR and SFR adversaries occupy the majority of clients or not. Extensive experimental results validate that PASS: (a) has the same level as the State-Of-The-Art method in mean square error against privacy leakage; (b) defends against AFR and SFR attacks in terms of a higher defense success rate, lower false positive rate, and higher F1-score; (c) is still effective where adversaries exceed 50%, with F1-score 89% against AFR attack and F1-score 87% against SFR attack. Note that PASS produces no negative effect on FL accuracy when there is no FR adversary.
- 23 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (1) , 1374-1384
PASS: A Parameter Audit-Based Secure and Fair Federated Learning Scheme Against Free-Rider Attack
Federated Learning (FL) as a secure distributed learning framework gains interests in Internet of Things (IoT) due to its capability of protecting the privacy of participant data. However, traditional FL systems are vulnerable to Free-Rider (FR) attacks, which causes unfairness, privacy leakage and inferior performance to FL systems. The prior defense mechanisms against FR attacks assumed that malicious clients (namely, adversaries) declare less than 50% of the total amount of clients. Moreover, they aimed for Anonymous FR (AFR) attacks and lost effectiveness in resisting Selfish FR (SFR) attacks. In this paper, we propose a Parameter Audit-based Secure and fair federated learning Scheme (PASS) against FR attack. PASS has the following key features: (a) prevent from privacy leakage with less accuracy loss; (b) be effective in countering both AFR and SFR attacks; (c) work well no matter whether AFR and SFR adversaries occupy the majority of clients or not. Extensive experimental results validate that PASS: (a) has the same level as the State-Of-The-Art method in mean square error against privacy leakage; (b) defends against AFR and SFR attacks in terms of a higher defense success rate, lower false positive rate, and higher F1-score; (c) is still effective where adversaries exceed 50%, with F1-score 89% against AFR attack and F1-score 87% against SFR attack. Note that PASS produces no negative effect on FL accuracy when there is no FR adversary.
- 23 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (1) , 1374-1384
PASS: A Parameter Audit-Based Secure and Fair Federated Learning Scheme Against Free-Rider Attack
Federated Learning (FL) as a secure distributed learning framework gains interests in Internet of Things (IoT) due to its capability of protecting the privacy of participant data. However, traditional FL systems are vulnerable to Free-Rider (FR) attacks, which causes unfairness, privacy leakage and inferior performance to FL systems. The prior defense mechanisms against FR attacks assumed that malicious clients (namely, adversaries) declare less than 50% of the total amount of clients. Moreover, they aimed for Anonymous FR (AFR) attacks and lost effectiveness in resisting Selfish FR (SFR) attacks. In this paper, we propose a Parameter Audit-based Secure and fair federated learning Scheme (PASS) against FR attack. PASS has the following key features: (a) prevent from privacy leakage with less accuracy loss; (b) be effective in countering both AFR and SFR attacks; (c) work well no matter whether AFR and SFR adversaries occupy the majority of clients or not. Extensive experimental results validate that PASS: (a) has the same level as the State-Of-The-Art method in mean square error against privacy leakage; (b) defends against AFR and SFR attacks in terms of a higher defense success rate, lower false positive rate, and higher F1-score; (c) is still effective where adversaries exceed 50%, with F1-score 89% against AFR attack and F1-score 87% against SFR attack. Note that PASS produces no negative effect on FL accuracy when there is no FR adversary.
- 15 November 2022
- Published by SPIE-Intl Soc Optical Eng
- Vol. 12448, 233-239
Drug screening methods based on deep learning
Drug screening is an important step in the development of new drugs. Through appropriate experimental methods and screening models, drugs with specific bioactivity can be transferred from laboratory research to clinical application. Nowadays, traditional efficiency of drug experimentation has been unable to meet the needs of our society. With the rapid development of computer technology, computer-assisted diagnosis and treatment have been gradually accepted and recognized by clinicians and patients. Drug screening process at the cellular level was studied in this paper. We not only compared the advantages and disadvantages of deep learning models and traditional machine learning methods, but also analyzed the performance of different deep learning models. First, Hela cells injected with different anti-stress drugs were divided into groups for experiment. G3BP, TIA-1 and the nucleus were labeled, respectively. The images were obtained using a single-photon microscope. Then, we distinguished the images of Hela cells treated with different drugs through visual observation, traditional machine learning (LBP/Gabor+SVM) and deep learning algorithms (VGGNet, GoogLeNet, ResNext and DenseNet), respectively. Experimental results showed that compared with visual observation, traditional machine learning and deep learning algorithms had better objectivity. Furthermore, deep learning models all had good classification performance. The highest average correct recognition rate was 92.97%, while that of the traditional method was only 80.93%. Therefore, drug screening methods based on deep learning could assist in screening the optimal treatment drugs, which help clinicians choose appropriate therapy.
- 15 November 2022
- Published by SPIE-Intl Soc Optical Eng
- Vol. 12448, 233-239
Drug screening methods based on deep learning
Drug screening is an important step in the development of new drugs. Through appropriate experimental methods and screening models, drugs with specific bioactivity can be transferred from laboratory research to clinical application. Nowadays, traditional efficiency of drug experimentation has been unable to meet the needs of our society. With the rapid development of computer technology, computer-assisted diagnosis and treatment have been gradually accepted and recognized by clinicians and patients. Drug screening process at the cellular level was studied in this paper. We not only compared the advantages and disadvantages of deep learning models and traditional machine learning methods, but also analyzed the performance of different deep learning models. First, Hela cells injected with different anti-stress drugs were divided into groups for experiment. G3BP, TIA-1 and the nucleus were labeled, respectively. The images were obtained using a single-photon microscope. Then, we distinguished the images of Hela cells treated with different drugs through visual observation, traditional machine learning (LBP/Gabor+SVM) and deep learning algorithms (VGGNet, GoogLeNet, ResNext and DenseNet), respectively. Experimental results showed that compared with visual observation, traditional machine learning and deep learning algorithms had better objectivity. Furthermore, deep learning models all had good classification performance. The highest average correct recognition rate was 92.97%, while that of the traditional method was only 80.93%. Therefore, drug screening methods based on deep learning could assist in screening the optimal treatment drugs, which help clinicians choose appropriate therapy.