Yun Wang

Dr. Yun Wang

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
1 April 2018 in The Annals of Thoracic Surgery
Verified
Journal Article
Closed access
HZHanlu Zhang
Hanlu Zhang
LCLongqi Chen
Longqi Chen
ZWZihao Wang
Zihao Wang
YZYu Zheng
Yu Zheng
YGYingcai Geng
Yingcai Geng
YWYun Wang
Yun Wang

11 March 2025 in Journal of Medical Internet Research
Verified
Journal Article
Open access
ZDZhang-Yi Dai
Zhang-Yi Dai
FWFu-Qiang Wang
Fu-Qiang Wang
CSCheng Shen
Cheng Shen
YJYan-Li Ji
Yan-Li Ji
ZLZhi-Yang Li
Zhi-Yang Li
QPQiang Pu
Qiang Pu
Background: Systematic reviews and meta-analyses rely on labor-intensive literature screening. While machine learning offers potential automation, its accuracy remains suboptimal. This raises the question of whether emerging large language models (LLMs) can provide a more accurate and efficient approach. Objective: This paper evaluates the sensitivity, specificity, and summary receiver operating characteristic (SROC) curve of LLM-assisted literature screening. Methods: We conducted a diagnostic study comparing the accuracy of LLM-assisted screening versus manual literature screening across 6 thoracic surgery meta-analyses. Manual screening by 2 investigators served as the reference standard. LLM-assisted screening was performed using ChatGPT-4o (OpenAI) and Claude-3.5 (Anthropic) sonnet, with discrepancies resolved by Gemini-1.5 pro (Google). In addition, 2 open-source, machine learning–based screening tools, ASReview (Utrecht University) and Abstrackr (Center for Evidence Synthesis in Health, Brown University School of Public Health), were also evaluated. We calculated sensitivity, specificity, and 95% CIs for the title and abstract, as well as full-text screening, generating pooled estimates and SROC curves. LLM prompts were revised based on a post hoc error analysis. Results: LLM-assisted full-text screening demonstrated high pooled sensitivity (0.87, 95% CI 0.77-0.99) and specificity (0.96, 95% CI 0.91-0.98), with the area under the curve (AUC) of 0.96 (95% CI 0.94-0.97). Title and abstract screening achieved a pooled sensitivity of 0.73 (95% CI 0.57-0.85) and specificity of 0.99 (95% CI 0.97-0.99), with an AUC of 0.97 (95% CI 0.96-0.99). Post hoc revisions improved sensitivity to 0.98 (95% CI 0.74-1.00) while maintaining high specificity (0.98, 95% CI 0.94-0.99). In comparison, the pooled sensitivity and specificity of ASReview tool-assisted screening were 0.58 (95% CI 0.53-0.64) and 0.97 (95% CI 0.91-0.99), respectively, with an AUC of 0.66 (95% CI 0.62-0.70). The pooled sensitivity and specificity of Abstrackr tool-assisted screening were 0.48 (95% CI 0.35-0.62) and 0.96 (95% CI 0.88-0.99), respectively, with an AUC of 0.78 (95% CI 0.74-0.82). A post hoc meta-analysis revealed comparable effect sizes between LLM-assisted and conventional screening. Conclusions: LLMs hold significant potential for streamlining literature screening in systematic reviews, reducing workload without sacrificing quality. Importantly, LLMs outperformed traditional machine learning-based tools (ASReview and Abstrackr) in both sensitivity and AUC values, suggesting that LLMs offer a more accurate and efficient approach to literature screening.

1 March 2018 in Journal of Thoracic Disease
Verified
Journal Article
Open access
PDF file
HZHanlu Zhang
Hanlu Zhang
LCLongqi Chen
Longqi Chen
YZYu Zheng
Yu Zheng
ZWZihao Wang
Zihao Wang
YGYingcai Geng
Yingcai Geng
YWYun Wang
Yun Wang
Background: Trans-subxiphoid thoracoscopic thymectomy is a promising procedure but technically demanding associated with ergonomic discomfort. In order to facilitate this complex procedure, the authors present a modified trans-subxiphoid thymectomy by the use of the Da Vinci robotic system. Methods: A 2-cm longitudinal incision for camera was made below the xiphoid process. Through this incision, the space among the posterior surface of the sternum, bilateral mediastinal pleura and diaphragm (extra pleural space) was enlarged blindly with a finger. Additional two operation holes were created below bilateral costal arches, and then two 8-mm robotic trocars were inserted into the extra pleural space. A third robotic instrument was placed in the 5th intercostal space after bilateral mediastinal pleura were dissected. The whole thymus was excised for patients with thymic tumor, and simultaneous resection of surrounding adipose tissues was performed for patients with myasthenia gravis. Clinical characteristics and early surgical outcomes of the patients were retrospectively collected and analyzed. Results: Between August 2016 and September 2017, 70 consecutive patients with myasthenia gravis or thymic neoplasm were successfully treated by the described surgical procedure. The median overall operative time was 115 (range, 60–150) min. The median hospital stay was 5 (range, 4–7) days. The median duration of chest tube was 3 (range, 1–8) days. The blood loss was minimal for all the patients. There was no mortality, conversion or postoperative complication during the postoperative and follow-up period. Conclusions: Based on our preliminary experience, the described technique of thymectomy is safe and feasible, and provides an optimal access for the robot camera and the instruments.

19 September 2019 in World Journal of Surgery
Verified
Journal Article
Closed access
PDF file
HZHanlu Zhang
Hanlu Zhang
ZWZihao Wang
Zihao Wang
YZYu Zheng
Yu Zheng
YGYingcai Geng
Yingcai Geng
FWFuqiang Wang
Fuqiang Wang
YWYun Wang
Yun Wang

1 July 2019 in United European Gastroenterology Journal
Verified
Journal Article
Open access
YZYan Zhu
Yan Zhu
JXJia-Xin Xu
Jia-Xin Xu
JCJing Cheng
Jing Cheng
ZZZhen Zhang
Zhen Zhang
BZBo-Qun Zhu
Bo-Qun Zhu
PZPing-Hong Zhou
Ping-Hong Zhou
Background and aims: Making an optimal and lasting submucosal cushion is critical for endoscopic submucosal dissection. The thermo‐sensitive binary hydrogels system composed of poloxamer 407 and poloxamer 188 might be an excellent submucosal injection solution considering the unique feature that it remains liquid at room temperature and becomes gelatinous after being injected in the submucosa of the digestive tract. The present study focuses on preparing the thermo‐sensitive binary hydrogels system and testing its capacity in mucosal lifting and its role in the endoscopic submucosal dissection procedure.Methods: Various concentrations of poloxamer 407 and poloxamer 188 were added to normal saline. The gelation temperature viscosity of the thermo‐sensitive binary hydrogels system was measured to choose the best formula. The thermo‐sensitive binary hydrogels system and normal saline were first compared in extracted porcine stomach. For in vivo study, the thermo‐sensitive binary hydrogels system and normal saline were compared for facilitating the endoscopic submucosal dissection procedure.Results: Among the 46 kinds of thermo‐sensitive binary hydrogels system, gelation temperatures of the thermo‐sensitive binary hydrogels system I (poloxamer 407/poloxamer 188, 17%/0.5%, w/w) and the thermo‐sensitive binary hydrogels system II (poloxamer 407/poloxamer 188, 18%/2%, w/w) were among the ideal range of gelation temperature. The injecting pressure in vitro study of thermo‐sensitive binary hydrogels system II was significantly higher than that of thermo‐sensitive binary hydrogels system I and normal saline (p < 0.001). Sixteen gastric endoscopic submucosal dissection procedures were performed in a porcine model. The initial volume of normal saline injection (13.88 ± 3.91 ml vs 5.88 ± 3.44 ml, p = 0.001) was significantly larger than the thermo‐sensitive binary hydrogels system group. The postoperative wound showed a significant difference in the two groups (p = 0.023) indicating that the thermo‐sensitive binary hydrogels system can create a cleaner wound.Conclusions: Considering the gelation temperature, viscosity, injection pressure, and the height of the mucosal elevation, the thermo‐sensitive binary hydrogels system I was the better submucosal injection solution.

28 May 2021 in Frontiers in Oncology
Verified
Journal Article
Open access
GQGuanghao Qiu
Guanghao Qiu
HZHanlu Zhang
Hanlu Zhang
FWFuqiang Wang
Fuqiang Wang
YZYu Zheng
Yu Zheng
ZWZihao Wang
Zihao Wang
YWYun Wang
Yun Wang
Background: Esophageal adenocarcinoma (EAC) is the most common kind of esophageal cancer. Age at diagnosis of advanced EAC is greater. Studies about practice patterns for elderly EAC patients with distant metastasis (DM) in stage IVB are limited. This retrospective, population-based study was conducted using data from the Surveillance, Epidemiology, and End Results (SEER) to evaluate 855 elderly EAC patients with DM in stage IVB from 2010 to 2015.Methods: 855 elderly EAC patients with DM in stage IVB between 2010 and 2015 were included in this study. Univariate and multivariate Cox-regression and Kaplan-Meier analyses were used to assess prognosis. These patients were classified to bone-only, brain-only, lung-only, liver-only, and multiple (patients with two or more organs in metastasis)-site group according to the site of metastasis. Overall survival (OS), cancer-specific survival (CSS), median survival time (MST), and survival rate (SR) were evaluated to analyze the survival outcomes.Results: The most common metastasis site was the liver among the single-organ metastasis population, followed by lung, bone, and brain. Compared with the bone-only group, the multiple-site group was associated with worst OS (HR: 1.037, 95% CI: 0.811–1.327, p = 0.770) and CSS (HR: 1.052, 95% CI: 0.816–1.357, p = 0.695). The multiple-site group also had the lowest MST in the population (MST: 2 months in OS and 3 months in CSS) and SR (6-month SR: 27.1% in OS, 29.9% in CSS, 1-year SR: 10.7% in OS, 12.0% in CSS, 3-year SR: 2.5% in OS, 2.8% in CSS). Compared to untreated patients (N) in the total population, other patients who were treated with surgery (S), radiotherapy (R), and chemotherapy (C) are beneficial for the prognosis (OS and CSS: p < 0.001).Conclusion: This population-based study was conducted to ascertain metastasis patterns and survival outcomes of EAC patients with DM in stage IVB. Elderly patients with multiple-site metastasis exhibited the worst OS and CSS among all the populations, and patients with bone-only metastasis had the worst OS and CSS among single-organ metastasis populations. Active treatment is beneficial for elderly EAC patients with DM in stage IVB, especially chemotherapy. This study also shows that more than one third of the patients had not received any therapy.