Ziyun Shen

Dr. Ziyun Shen


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
23 January 2024 in International Journal of Surgery
Verified
Journal Article
Open access
CTChaoyang Tong
Chaoyang Tong
XDXinwei Du
Xinwei Du
YCYancheng Chen
Yancheng Chen
KZKan Zhang
Kan Zhang
MSMengqin Mm Shan
Mengqin Mm Shan
JZJijian Zheng
Jijian Zheng
Background: Major adverse postoperative outcomes (APOs) can greatly affect mortality, hospital stay, care management and planning, and quality of life. This study aimed to evaluate the performance of five machine learning (ML) algorithms for predicting four major APOs after pediatric congenital heart surgery and their clinically meaningful model interpretations. Methods: Between August 2014 and December 2021, 23 000 consecutive pediatric patients receiving congenital heart surgery were enrolled. Based on the split date of 1 January 2019, the authors selected 13 927 participants for the training cohort, and 9073 participants for the testing cohort. Four predefined major APOs including low cardiac output syndrome (LCOS), pneumonia, renal failure, and deep venous thrombosis (DVT) were investigated. Thirty-nine clinical and laboratory features were inputted in five ML models: light gradient boosting machine (LightGBM), logistic regression (LR), support vector machine, random forest, and CatBoost. The performance and interpretations of ML models were evaluated using the area under the receiver operating characteristic curve (AUC) and Shapley Additive Explanations (SHAP). Results: In the training cohort, CatBoost algorithms outperformed others with the mean AUCs of 0.908 for LCOS and 0.957 for renal failure, while LightGBM and LR achieved the best mean AUCs of 0.886 for pneumonia and 0.942 for DVT, respectively. In the testing cohort, the best-performing ML model for each major APOs with the following mean AUCs: LCOS (LightGBM), 0.893 (95% CI: 0.884–0.895); pneumonia (LR), 0.929 (95% CI: 0.926–0.931); renal failure (LightGBM), 0.963 (95% CI: 0.947–0.979), and DVT (LightGBM), 0.970 (95% CI: 0.953–0.982). The performance of ML models using only clinical variables was slightly lower than those using combined data, with the mean AUCs of 0.873 for LCOS, 0.894 for pneumonia, 0.953 for renal failure, and 0.933 for DVT. The SHAP showed that mechanical ventilation time was the most important contributor of four major APOs. Conclusions: In pediatric congenital heart surgery, the established ML model can accurately predict the risk of four major APOs, providing reliable interpretations for high-risk contributor identification and informed clinical decisions-making.

1 June 2022 in International Journal of Surgery
Verified
Journal Article
Closed access
ZSZiyun Shen
Ziyun Shen
HCHaoda Chen
Haoda Chen
WWWeishen Wang
Weishen Wang
WXWei Xu
Wei Xu
YZYiran Zhou
Yiran Zhou
BSBaiyong Shen
Baiyong Shen
Clinically relevant postoperative pancreatic fistula (CR-POPF) remains the major cause of morbidity following pancreaticoduodenectomy (PD). Several model score systems such as the Fistula Risk Score (FRS) have been developed to predict CR-POPF using preoperative and intraoperative data. Machine learning (ML) algorithms are increasingly applied in the medical field and they could be used to assess the risk of CR-POPF, identify clinically meaningful data and guide drain removal. Data from consecutive patients who underwent PD between January 1, 2010 and March 31, 2021 at a single high-volume center was collected retrospectively in this study. Demographics, clinical features, intraoperative parameters, and laboratory values were used to conduct the ML model. Four different ML algorithms (CatBoost, lightGBM, XGBoost and Random Forest) were used to train this model with cross-validation. A total of 2421 patients with 62 clinical parameters were enrolled in this ML model. The majority of patients (76.3%) underwent open PD while others underwent robot-assisted PD. CR-POPF occurred in 424 (17.5%) patients. The CatBoost algorithm outperformed other algorithms with a mean area under the receiver operating characteristic curve (AUC) of 0.81 (95% confidence interval: 0.80-0.82) from the 5-fold cross-validation procedure. In the test dataset, the CatBoost algorithm also achieved the best mean-AUC of 0.83. The most important value was mean drain fluid amylase (DFA) in the first seven postoperative days (POD). The performance of models that used only preoperative data and intraoperative data was marginally lower than that of models that used combined data. Our ML algorithms could be applied as early diagnostic tools for CR-POPF in patients who underwent PD. Such real-time clinical decision support tools can identify patients with a high risk of CR-POPF, help in developing the perioperative management plan and guide the optimal timing of drain removal.

1 June 2021 in International Journal of Surgery
Verified
Journal Article
Closed access
HCHaoda Chen
Haoda Chen
ZSZiyun Shen
Ziyun Shen
XYXiayang Ying
Xiayang Ying
YWYuanchi Weng
Yuanchi Weng
YJYu Jiang
Yu Jiang
BSBaiyong Shen
Baiyong Shen
Several studies have suggested an association between visceral obesity and adverse perioperative outcomes in pancreatic surgery. However, no study has reported the impact of visceral obesity on robotic pancreatic surgery. This study aimed to assess the impact of preoperative visceral obesity on clinically relevant postoperative pancreatic fistula (CR-POPF) in pancreatic ductal adenocarcinoma (PDAC) patients following robotic distal pancreatectomy (RDP) or open distal pancreatectomy (ODP). A retrospective analysis was performed on all patients who consecutively underwent RDP or ODP for PDAC. The visceral adipose tissue was measured on preoperative computed tomography (CT) images at the L3 vertebra level. A 1:1 propensity score matching method was used in the visceral obesity group and the nonvisceral obesity group to minimize the bias between RDP and ODP. Between December 2011 and December 2018, a total of 445 patients were included. Visceral obesity (n = 219) was found to be associated with higher estimated blood loss (p = 0.033), a higher CR-POPF rate (p = 0.001), delayed drain removal (p = 0.005) and a longer length of stay (p = 0.033). In multivariable analysis, visceral obesity was an independent risk factor for CR-POPF (OR: 1.69; 95% CI 1.07-2.67, p = 0.024). Among patients without visceral obesity, the incidence of CR-POPF was lower among RDP than among ODP patients (11.27% vs. 23.87%, p = 0.028), and the difference remained after propensity score matching (9.52% vs. 26.98%, p = 0.011). However, there was no significant difference in the CR-POPF rate between RDP and ODP for visceral obesity patients. In nonvisceral obesity patients, RDP had better perioperative outcomes than ODP, with a significantly lower CR-POPF rate. However, in visceral obesity patients, RDP showed equivalent CR-POPF rate when compared to ODP.

2 August 2024 in European Journal of Cardio-Thoracic Surgery
Verified
Journal Article
Closed access
PDF file
MTMeixin Teng
Meixin Teng
CYChengxiang Yi
Chengxiang Yi
WYWeiguang Yang
Weiguang Yang
JZJing Zhang
Jing Zhang
WYWangchao Yao
Wangchao Yao
PZPeng Zhang
Peng Zhang
Neoadjuvant therapy has gained widespread acceptance as the standard modality for locally advanced non-small-cell lung cancer (NSCLC). However, the clinical benefit of sleeve lobectomy (SL) or pneumonectomy (PN) following neoadjuvant therapy remains controversial. The clinical and pathological characteristics of NSCLC patients who underwent SL or PN after neoadjuvant therapy at a high-volume single centre between December 2019 and March 2023 were retrospectively collected. The SL group was matched 4:1 with the PN group by propensity score matching (PSM). The surgical outcomes were systematically collected and analysed. During a 5-year study period, the majority of patients (175 out of 215, 81.4%) underwent the SL procedure, while 40 patients (18.6%) underwent PN. Following PSM, the SL group exhibited lower postoperative arrythmia (4.8% vs 26.9%, P < 0.001), lower 30-day mortality (1.0% vs 7.7%, P = 0.046) and a shorter length of postoperative hospital stay (6.0 days vs 10.0 days, P < 0.001), compared to the PN group. In addition, no significant difference was observed between the two groups in terms of disease-free survival or overall survival (P = 0.977 and P = 0.913, respectively). SL stands as a safe and feasible option for patients with centrally located NSCLC who have undergone neoadjuvant therapy, in comparison to PN. This finding suggests that SL remains the preferable choice when feasible in the context of the widespread utilization of neoadjuvant therapy.

1 April 2025 in Journal of Thoracic Disease
Verified
Journal Article
Open access
PDF file
SHShiqi Hu
Shiqi Hu
XZXiaoping Zeng
Xiaoping Zeng
SWShiyue Wan
Shiyue Wan
SWSuyu Wang
Suyu Wang
QZQianxin Zhou
Qianxin Zhou
PZPeng Zhang
Peng Zhang
Original Article Comparison of robot-assisted and video-assisted thoracic sleeve lobectomy in non-small cell lung cancer: insights from a high-volume center Shiqi Hu, Xiaoping Zeng, Shiyue Wan, Suyu Wang, Qianxin Zhou, Ziyun Shen, Jing Zhang, Peng Zhang

9 March 2021 in Clinica Chimica Acta
Verified
Journal Article
Closed access
ZSZiyun Shen
Ziyun Shen
ZXZhiwei Xu
Zhiwei Xu
WWWeishen Wang
Weishen Wang
WXWei Xu
Wei Xu
YZYiran Zhou
Yiran Zhou
BSBaiyong Shen
Baiyong Shen