Liyan Sun

Dr. Liyan Sun

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 April 2022 in IEEE Access
Verified
Journal Article
Open access
LLLina Li
Lina Li
MLMinghan Liu
Minghan Liu
LSLiyan Sun
Liyan Sun
YLYupeng Li
Yupeng Li
NLNianfeng Li
Nianfeng Li
Recognizing students’ behaviors in classes based on videos of their activity plays an important role in improving teaching quality and paying attention to the healthy growth of students. The existing student behavior recognition methods mainly focus on the behavior of the single student and have low performance and efficiency. To recognize the behaviors of multiple students in the classroom at the same time, we propose a fast and effective solution, called ET-YOLOv5s, which is an improved YOLOv5s with ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) and a tiny object detection module in this paper. First, the ESRGAN is adopted to generate high-definition images from original images of real classroom environments in colleges to improve the recognition accuracy. Then a tiny object detection module is added to the YOLOv5s to detect the smaller objects on high-definition images, such as phones, books, and students sitting at the back of the classroom. Finally, the experimental results show that the proposed solution can detect 11 kinds of students’ behaviors in classrooms, including playing mobile phones, having class, sitting with hands on face, sitting turn right, sitting turn left, bowing, sleeping, and so on. Compared with YOLOv4, YOLOv5s, and other tiny target detection algorithms, it has better detection performance and can effectively deal with the behavior recognition of multiple students.

22 August 2018 in BMC Bioinformatics
Verified
Journal Article
Open access
PDF file
RWRongquan Wang
Rongquan Wang
GLGuixia Liu
Guixia Liu
CWCaixia Wang
Caixia Wang
LSLingtao Su
Lingtao Su
LSLiyan Sun
Liyan Sun
In recent decades, detecting protein complexes (PCs) from protein-protein interaction networks (PPINs) has been an active area of research. There are a large number of excellent graph clustering methods that work very well for identifying PCs. However, most of existing methods usually overlook the inherent core-attachment organization of PCs. Therefore, these methods have three major limitations we should concern. Firstly, many methods have ignored the importance of selecting seed, especially without considering the impact of overlapping nodes as seed nodes. Thus, there may be false predictions. Secondly, PCs are generally supposed to be dense subgraphs. However, the subgraphs with high local modularity structure usually correspond to PCs. Thirdly, a number of available methods lack handling noise mechanism, and miss some peripheral proteins. In summary, all these challenging issues are very important for predicting more biological overlapping PCs. In this paper, to overcome these weaknesses, we propose a clustering method by core-attachment and local modularity structure, named CALM, to detect overlapping PCs from weighted PPINs with noises. Firstly, we identify overlapping nodes and seed nodes. Secondly, for a node, we calculate the support function between a node and a cluster. In CALM, a cluster which initially consists of only a seed node, is extended by adding its direct neighboring nodes recursively according to the support function, until this cluster forms a locally optimal modularity subgraph. Thirdly, we repeat this process for the remaining seed nodes. Finally, merging and removing procedures are carried out to obtain final predicted clusters. The experimental results show that CALM outperforms other classical methods, and achieves ideal overall performance. Furthermore, CALM can match more complexes with a higher accuracy and provide a better one-to-one mapping with reference complexes in all test datasets. Additionally, CALM is robust against the high rate of noise PPIN. By considering core-attachment and local modularity structure, CALM could detect PCs much more effectively than some representative methods. In short, CALM could potentially identify previous undiscovered overlapping PCs with various density and high modularity.

7 August 2023 in IEEE Access
Verified
Journal Article
Open access
NLNianfeng Li
Nianfeng Li
XSXIANGFENG SHEN
XIANGFENG SHEN
LSLiyan Sun
Liyan Sun
ZXZhiguo Xiao
Zhiguo Xiao
TDTianjiao Ding
Tianjiao Ding
XLXinhang Li
Xinhang Li
Since the position of the classroom surveillance camera is not fixed, the angle of the face captured through the surveillance video is also different. The deep learning-based face verification model has achieved good results in controlled environments, but there is still much room for improvement in the face verification ability in uncontrolled environments. The performance of the model depends not only on the results of the network but also on the quality and diversity of the dataset. The current Asian face dataset in an uncontrolled environment is insufficient; for this reason, this paper constructs a Chinese face dataset (UCEC-Face) in an uncontrolled classroom environment, which is collected by 35 real classroom surveillance videos. The UCEU dataset contains 7395 images of 130 subjects, including 44 males and 86 females. To verify that there is still room for improving the performance of existing face verification models for Asian face verification, we further utilize four models such as OpenFace and ArcFace for face verification, as well as the VGG-Face model for gender, expression, and age recognition on the UCEC-Face. The experimental results show that the UCEC-Face constructed in this paper is more challenging and difficult to verify in face verification tasks because it is closer to the real environment, and the best results obtained on the existing models only reach 69.7%, which is largely below the average accuracy of the identification results of other datasets.

7 August 2019 in BMC Genomics
Verified
Journal Article
Open access
PDF file
RWRongquan Wang
Rongquan Wang
CWCaixia Wang
Caixia Wang
LSLiyan Sun
Liyan Sun
GLGuixia Liu
Guixia Liu
The detection of protein complexes is of great significance for researching mechanisms underlying complex diseases and developing new drugs. Thus, various computational algorithms have been proposed for protein complex detection. However, most of these methods are based on only topological information and are sensitive to the reliability of interactions. As a result, their performance is affected by false-positive interactions in PPINs. Moreover, these methods consider only density and modularity and ignore protein complexes with various densities and modularities. To address these challenges, we propose an algorithm to exploit protein complexes in PPINs by a Seed-Extended algorithm based on Density and Modularity with Topological structure and GO annotations, named SE-DMTG to improve the accuracy of protein complex detection. First, we use common neighbors and GO annotations to construct a weighted PPIN. Second, we define a new seed selection strategy to select seed nodes. Third, we design a new fitness function to detect protein complexes with various densities and modularities. We compare the performance of SE-DMTG with that of thirteen state-of-the-art algorithms on several real datasets. The experimental results show that SE-DMTG not only outperforms some classical algorithms in yeast PPINs in terms of the F-measure and Jaccard but also achieves an ideal performance in terms of functional enrichment. Furthermore, we apply SE-DMTG to PPINs of several other species and demonstrate the outstanding accuracy and matching ratio in detecting protein complexes compared with other algorithms.

13 December 2019 in Current Bioinformatics
Verified
Journal Article
Closed access
PDF file
LSLiyan Sun
Liyan Sun
GLGuixia Liu
Guixia Liu
LSLingtao Su
Lingtao Su
RWRongquan Wang
Rongquan Wang
Background:: Genome-Wide Association Study (GWAS) plays a very important role in identifying the causes of a disease. Because most of the existing methods for genetic-interaction detection in GWAS are designed for a single-correlation model, their performances vary considerably for different disease models. These methods usually have high computation cost and low accuracy. Method:: We present a new multi-objective heuristic optimization methodology named HSMMGKG for detecting genetic interactions. In HS-MMGKG, we use harmony search with five objective functions to improve the efficiency and accuracy. A new strategy based on p-value and MDR is adopted to generate more reasonable results. The Boolean representation in BOOST is modified to calculate the five functions rapidly. These strategies take less time complexity and have higher accuracy while detecting the potential models. Results:: We compared HS-MMGKG with CSE, MACOED and FHSA-SED using 26 simulated datasets. The experimental results demonstrate that our method outperforms others in accuracy and computation time. Our method has identified many two-locus SNP combinations that are associated with seven diseases in WTCCC dataset. Some of the SNPs have direct evidence in CTD database. The results may be helpful to further explain the pathogenesis. Conclusion:: It is anticipated that our proposed algorithm could be used in GWAS which is helpful in understanding disease mechanism, diagnosis and prognosis.

19 June 2025 in IEEE Access
Verified
Journal Article
Open access
LSLiyan Sun
Liyan Sun
琳郑琳宣 郑
琳宣 郑
ZXZhiguo Xiao
Zhiguo Xiao
XYXin Yi
Xin Yi
LJLinqing Jiang
Linqing Jiang
Brain tumors are among the deadliest malignant neoplasms worldwide, and early detection is critical for improving treatment outcomes and patient prognosis. Magnetic resonance imaging (MRI) plays a central role in the clinical diagnosis of brain tumors due to its exceptional ability to visualize brain structures and detect tumors. However, traditional MRI detection models for brain tumors often struggle to balance high accuracy with a lightweight design. To address this challenge, this study proposes a novel model—STAR-YOLO—based on the optimization of YOLOv8n. The model refines the backbone network and enhances the C2f module within the neck network. It also integrates SimAM, a parameter-free attention mechanism, and utilizes ShapeIoU as the localization loss function. These innovations enable STAR-YOLO to achieve both high accuracy and a lightweight design. Experimental results on the Br35H public dataset show that STAR-YOLO outperforms existing models, achieving outstanding scores of 0.937 in Precision, 0.852 in Recall, and 0.642 in mAP50-95. Additionally, the model contains only 2.01 million parameters and has a computational complexity of 6.1 GFLOPs, significantly lower than other models. STAR-YOLO accomplishes a lightweight design while guaranteeing high detection accuracy, prominently demonstrating its immense potential in the diagnosis of clinical brain tumors, particularly in circumstances with constrained computing resources.