Bin Sun

Bin Sun


Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
2 April 2024 in Frontiers in Endocrinology
Verified
Journal Article
Open access
PDF file
CYChunfang Yan
Chunfang Yan
ZZZhaoxia Zhang
Zhaoxia Zhang
GZGuanghua Zhang
Guanghua Zhang
HLHan Liu
Han Liu
RZRuiqi Zhang
Ruiqi Zhang
BSBin Sun
Bin Sun
Background: Thyroid-associated ophthalmopathy (TAO) is the most prevalent autoimmune orbital condition, significantly impacting patients’ appearance and quality of life. Early and accurate identification of active TAO along with timely treatment can enhance prognosis and reduce the occurrence of severe cases. Although the Clinical Activity Score (CAS) serves as an effective assessment system for TAO, it is susceptible to assessor experience bias. This study aimed to develop an ensemble deep learning system that combines anterior segment slit-lamp photographs of patients with facial images to simulate expert assessment of TAO.Method: The study included 156 patients with TAO who underwent detailed diagnosis and treatment at Shanxi Eye Hospital Affiliated to Shanxi Medical University from May 2020 to September 2023. Anterior segment slit-lamp photographs and facial images were used as different modalities and analyzed from multiple perspectives. Two ophthalmologists with more than 10 years of clinical experience independently determined the reference CAS for each image. An ensemble deep learning model based on the residual network was constructed under supervised learning to predict five key inflammatory signs (redness of the eyelids and conjunctiva, and swelling of the eyelids, conjunctiva, and caruncle or plica) associated with TAO, and to integrate these objective signs with two subjective symptoms (spontaneous retrobulbar pain and pain on attempted upward or downward gaze) in order to assess TAO activity.Results: The proposed model achieved 0.906 accuracy, 0.833 specificity, 0.906 precision, 0.906 recall, and 0.906 F1-score in active TAO diagnosis, demonstrating advanced performance in predicting CAS and TAO activity signs compared to conventional single-view unimodal approaches. The integration of multiple views and modalities, encompassing both anterior segment slit-lamp photographs and facial images, significantly improved the prediction accuracy of the model for TAO activity and CAS.Conclusion: The ensemble multi-view multimodal deep learning system developed in this study can more accurately assess the clinical activity of TAO than traditional methods that solely rely on facial images. This innovative approach is intended to enhance the efficiency of TAO activity assessment, providing a novel means for its comprehensive, early, and precise evaluation.

9 April 2021 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
YZYuxuan Zhai
Yuxuan Zhai
GZGuanghua Zhang
Guanghua Zhang
LZLongsheng Zheng
Longsheng Zheng
GYGuangqian Yang
Guangqian Yang
KZKe Zhao
Ke Zhao
ZWZhao Wang
Zhao Wang
Cataract causes more than half of all blindness worldwide. The most effective treatment is surgery, where cataract is often replaced by intraocular lens (IOL). Beyond saving vision, toric IOL implantation is becoming increasingly popular to correct corneal astigmatism. It is important to precisely position and align the axis of IOL during surgery to achieve optimal post-operative astigmatism correction. Comparing with conventional manual marking, automated markerless IOL alignment can be faster, more accurate and non-invasive. Here we propose a framework for computer-assisted intraoperative IOL positioning and alignment based on detection and tracking. Firstly, the iris boundary was segmented and the eye center was determined. A statistical sampling method was developed to segment iris and generate training labels, and both conventional algorithms and deep convolutional neural network (CNN) methods were evaluated. Then, regions of interests (ROIs) containing high density of scleral capillaries were used for tracking eye rotations. Both correlation filter and CNN methods were evaluated for tracking. Cumulative errors during long-term tracking were corrected using a reference image. Validation studies against manual labeling using 7 clinical cataract surgical videos demonstrated that the proposed algorithm achieved an average position error around 0.2 mm, an axis alignment error of < 1 $^{\circ}$ , and a frame rate of > 25 FPS, and can be potentially used intraoperatively for markerless IOL positioning and alignment during cataract surgery.

11 November 2022 in Disease Markers
Verified
Journal Article
Open access
PDF file
GZGuanghua Zhang
Guanghua Zhang
BSBin Sun
Bin Sun
ZZZhaoxia Zhang
Zhaoxia Zhang
SWShiyu Wu
Shiyu Wu
GZGuangping Zhuo
Guangping Zhuo
WYWeihua Yang
Weihua Yang
Retinal vein occlusion (RVO) is one of the most common retinal vascular diseases leading to vision loss if not diagnosed and treated in time. RVO can be classified into two types: CRVO (blockage of the main retinal veins) and BRVO (blockage of one of the smaller branch veins). Automated diagnosis of RVO can improve clinical workflow and optimize treatment strategies. However, to the best of our knowledge, there are few reported methods for automated identification of different RVO types. In this study, we propose a new hypermixed convolutional neural network (CNN) model, namely, the VGG-CAM network, that can classify the two types of RVOs based on retinal fundus images and detect lesion areas using an unsupervised learning method. The image data used in this study is collected and labeled by three senior ophthalmologists in Shanxi Eye Hospital, China. The proposed network is validated to accurately classify RVO diseases and detect lesions. It can potentially assist in further investigating the association between RVO and brain vascular diseases and evaluating the optimal treatments for RVO.

26 July 2022 in Frontiers in Cellular and Infection Microbiology
Verified
Journal Article
Open access
XJXuan Ji
Xuan Ji
KDKui Dong
Kui Dong
JPJi Pu
Ji Pu
JYJing Yang
Jing Yang
ZZZhaoxia Zhang
Zhaoxia Zhang
BSBin Sun
Bin Sun
Purpose: Thyroid-associated ophthalmopathy (TAO) is a chronic autoimmune disease. In this study, high-throughput sequencing was used to investigate the diversity and composition of the ocular microbiota in patients with TAO.Methods: Patients with TAO did not receive treatment for the disease and did not have exposed keratitis. Patients with TAO (TAO group) and healthy individuals (control group) were compared. All samples were swabbed at the conjunctival vault of the lower eyelid. The V3 to V4 region of the 16S rDNA was amplified using polymerase chain reaction and sequenced on the Illumina HiSeq 2500 Sequencing Platform. Statistical analysis was performed to analyze the differences between the groups and the correlation between ocular surface microbiota and the disease. The ocular surface microbiota of patients and healthy individuals were cultured.Results: The ocular surface microbiota structure of TAO patients changed significantly. The average relative abundance of Bacillus and Brevundimonas increased significantly in the TAO group. Corynebacterium had a significantly decreased relative abundance (P<0.05). Paracoccus, Haemophilus, Lactobacillus, and Bifidobacterium were positively correlated with the severity of clinical manifestations or disease activity (P<0.05). Bacillus cereus and other opportunistic pathogens were obtained by culture from TAO patients.Conclusions: This study found that the composition of ocular microbiota in patients with TAO was significantly different from that in healthy individuals. The ocular surface opportunistic pathogens, such as Bacillus, Brevundimonas, Paracoccus, and Haemophilus in TAO patients, increase the potential risk of ocular surface infection. The findings of this study provide a new avenue of research into the mechanism of ocular surface in TAO patients.

2 September 2022 in Frontiers in Microbiology
Verified
Journal Article
Open access
KDKui Dong
Kui Dong
JPJi Pu
Ji Pu
JYJing Yang
Jing Yang
GZGuohong Zhou
Guohong Zhou
XJXuan Ji
Xuan Ji
JXJianguo Xu
Jianguo Xu
Objectives: To characterize the healthy ocular surface microbiota at the species level, including cultured and uncultured taxa.Methods: We integrated the metataxonomic method with culturomics and genome sequencing analysis of selected isolated strains to better illustrate the taxonomic structure of the ocular surface microbiota. The metataxonomics used the full-length 16S rRNA gene sequences and the operational phylogenetic unit strategy, which can precisely identify the cultured and uncultured or potentially new taxa to species level based on the phylogenetic tree constructed.Results: We detected 1,731 operational phylogenetic units (OPUs) in 196 healthy eyes from 128 people, affiliated to 796 cultured species, 784 potentially new species, and 151 potentially new higher taxa. The microbiota for each eye had 49.17 ± 35.66 OPUs. Of the 796 cultured species, 170 (21.36%) had previously caused clinical infections. Based on where they were initially isolated, the ocular surface microbiota mainly came from human body sites (34.55%), the environment (36.93%), plants (9.05%), animals (4.90%), and others; 428 strains were isolated from 20 eyes, affiliated to 42 species, and had come from the environment (33.33%) and the skin (16.67%). Of these, 47.62% had previously caused clinical infections. Genome analysis of 73 isolators revealed that 68.5% of them carried antibiotic resistance genes. The most frequently isolated genera, namely Staphylococcus, Streptococcus, and Moraxella, had an average of 5.30, four, and three resistance genes per strain, respectively.Discussion: The study found that the ocular surface microbiota mainly came from the environment, plants, animals, food, and human body sites such as the skin, oral cavity, upper respiratory tract, etc. No core member of ocular surface microbiota was detected at the species level. The human eyes were invaded and colonized by bacteria from the exposed environment, some of which were capable of causing infections in humans and carried antibiotic resistance genes. Preventive measures should be developed to protect our eyes from danger.

1 December 2021 in Translational Vision Science & Technology
Verified
Journal Article
Open access
PDF file
KDKui Dong
Kui Dong
ZKZhiMing Kang
ZhiMing Kang
XJXuan Ji
Xuan Ji
XZXinxin Zhang
Xinxin Zhang
PCPeiNi Cheng
PeiNi Cheng
BSBin Sun
Bin Sun
Purpose: Pseudomonas aeruginosa is the most common bacteria causing endophthalmitis after cataract surgery. Vitreous fluid culture and molecular studies are commonly used in clinical diagnoses, but have disadvantages, such as a long culture cycle and low detection sensitivity. Here, we report a loop-mediated isothermal amplification (LAMP) method combined with the nanoparticles-lateral flow biosensor (LFB) method for rapid and specific detection of P. aeruginosa. Methods: A set of six primers was designed to target the OprL gene of P. aeruginosa. Genomic DNA extracted from several gram-negative and gram-positive bacteria was used to determine the sensitivity and specificity of the analysis. LAMP reactions were conducted at 65 °C for 50 minutes, and results were reported using the LFB method. Results: The DNA template of P. aeruginosa was specifically recognized by the P. aeruginosa-LAMP-LFB (PA-LAMP-LFB) method as no cross reactions were observed for non–P. aeruginosa templates. The analytical sensitivity of our assay was 100 fg per test for the pure cultured DNA template, and the result obtained using the LFB was consistent with that of colorimetric indicator detection. The whole test could be completed within 1h. This method was used to detect P. aeruginosa, Staphylococcus aureus, and Klebsiella pneumoniae; only P. aeruginosa was positive. The positive rates of P. aeruginosa detected by a traditional culture method, the LAMP-LFB method, and the fluorescence quantitative polymerase chain reaction method were 17.7%, 17.7%, and 13.3%, respectively. Conclusions: The P. aeruginosa-LAMP-LFB method established here is a rapid, specific, and sensitive method for the detection of P. aeruginosa, which can be widely used.