Shun Hu
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
Shun Hu has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Fei Li
Department of General Surgery, Xuanwu Hospital, Capital Medical University
2 shared publications - Feng Cao
Department of General Surgery, Xuanwu Hospital, Capital Medical University, Beijing, China
2 shared publications - Yulin Guo
Clinical Center of Acute Pancreatitis, Capital Medical University, Beijing, People’s Republic of China
2 shared publications - Ang Li
Department of General Surgery, Xuanwu Hospital, Capital Medical University, Beijing 100053, China
1 shared publication - Anjia Yang
Jinan University
1 shared publication Caiqin Dong
1 shared publication
Latest Publications
- Open Access
Fusion: Efficient and Secure Inference Resilient to Malicious Servers
In secure machine learning inference, most of the schemes assume that the server is semi-honest (honestly following the protocol but attempting to infer additional information). However, the server may be malicious (e.g., using a low-quality model or deviating from the protocol) in the real world. Although a few studies have considered a malicious server that deviates from the protocol, they ignore the verification of model accuracy (where the malicious server uses a low-quality model) meanwhile preserving the privacy of both the server's model and the client's inputs. To address these issues, we propose \textit{Fusion}, where the client mixes the public samples (which have known query results) with their own samples to be queried as the inputs of multi-party computation to jointly perform the secure inference. Since a server that uses a low-quality model or deviates from the protocol can only produce results that can be easily identified by the client, \textit{Fusion} forces the server to behave honestly, thereby addressing all those aforementioned issues without leveraging expensive cryptographic techniques. Our evaluation indicates that \textit{Fusion} is 48.06$\times$ faster and uses 30.90$\times$ less communication than the existing maliciously secure inference protocol (which currently does not support the verification of the model accuracy). In addition, to show the scalability, we conduct ImageNet-scale inference on the practical ResNet50 model and it costs 8.678 minutes and 10.117 GiB of communication in a WAN setting, which is 1.18$\times$ faster and has 2.64$\times$ less communication than those of the semi-honest protocol.
Open AccessFusion: Efficient and Secure Inference Resilient to Malicious Servers
In secure machine learning inference, most of the schemes assume that the server is semi-honest (honestly following the protocol but attempting to infer additional information). However, the server may be malicious (e.g., using a low-quality model or deviating from the protocol) in the real world. Although a few studies have considered a malicious server that deviates from the protocol, they ignore the verification of model accuracy (where the malicious server uses a low-quality model) meanwhile preserving the privacy of both the server's model and the client's inputs. To address these issues, we propose \textit{Fusion}, where the client mixes the public samples (which have known query results) with their own samples to be queried as the inputs of multi-party computation to jointly perform the secure inference. Since a server that uses a low-quality model or deviates from the protocol can only produce results that can be easily identified by the client, \textit{Fusion} forces the server to behave honestly, thereby addressing all those aforementioned issues without leveraging expensive cryptographic techniques. Our evaluation indicates that \textit{Fusion} is 48.06$\times$ faster and uses 30.90$\times$ less communication than the existing maliciously secure inference protocol (which currently does not support the verification of the model accuracy). In addition, to show the scalability, we conduct ImageNet-scale inference on the practical ResNet50 model and it costs 8.678 minutes and 10.117 GiB of communication in a WAN setting, which is 1.18$\times$ faster and has 2.64$\times$ less communication than those of the semi-honest protocol.
- Open Access
- 28 October 2022
- Published by Wiley in Emergency Medicine International
- Vol. 2022, 1-8
COVID-19 Infection Deteriorates the Clinical Condition and Outcomes of Acute Pancreatitis: A Meta-Analysis
Backgrounds. The novel coronavirus disease 2019 (COVID-19) has caused a global pandemic. Pancreatic injuries have been reported in COVID-19 patients. The present meta-analysis was conducted to compare the morbidity and outcomes of AP between COVID-19 positive and negative patients. Methods. Databases including Cochrane Library, PubMed, and EMBASE were systematically searched (until July 3rd 2022). Studies with English abstracts comparing the severity and outcomes of AP between COVID-19 positive and negative patients were included. Mean differences or odds ratios with a 95% confidence interval were employed for assess variables. Risk of publication bias was assessed with funnel plots. Results. Data from 7 studies with a total of 2816 AP patients were included. COVID-19 positive was associated with higher incidences of pancreatic necrosis (OR = 1.65; 95% CI: 1.13 to 2.42, = 0.01; = 0.82 for heterogeneity) and persistent organ failure (OR = 6.87; 95% CI: 2.37 to 19.98, = 0.0004; = 0.12 for heterogeneity), especially cardiovascular failure (OR = 2.92; 95% CI: 1.66 to 5.14, = 0.0002; = 0.58 for heterogeneity) and acute respiratory distress syndrome (ARDS) or acute lung injury (ALI) (OR = 3.03; 95% CI: 2.09 to 4.39, < 0.00001; = 0.20 for heterogeneity). COVID-19 infection induced a higher level of CRP (MD = 0.40; 95% CI: 0.16 to 0.64, = 0.001; < 0.00001 for heterogeneity) as well as coagulation disorders involving platelets, prothrombin time, activated partial thromboplastin time, and D-dimer (all < 0.05). During hospitalization, COVID-19 positive was associated with higher ICU admission rate (OR = 2.76; 95% CI: 1.98 to 3.85 < 0.00001; = 0.47 for heterogeneity). COVID-19 positive AP was associated with a higher mortality rate (OR = 3.70; 95% CI: 2.60 to 5.25, < 0.00001; = 0.12 for heterogeneity). Discussion. The number of included studies is limited and none is RCT, thus the risks of publication and selective bias could not be ignored. COVID-19 deteriorated the severity and clinical outcomes of AP, with a high incidence of morbidity and mortality.
- 28 October 2022
- Published by Wiley in Emergency Medicine International
- Vol. 2022, 1-8
Open AccessCOVID-19 Infection Deteriorates the Clinical Condition and Outcomes of Acute Pancreatitis: A Meta-Analysis
Backgrounds. The novel coronavirus disease 2019 (COVID-19) has caused a global pandemic. Pancreatic injuries have been reported in COVID-19 patients. The present meta-analysis was conducted to compare the morbidity and outcomes of AP between COVID-19 positive and negative patients. Methods. Databases including Cochrane Library, PubMed, and EMBASE were systematically searched (until July 3rd 2022). Studies with English abstracts comparing the severity and outcomes of AP between COVID-19 positive and negative patients were included. Mean differences or odds ratios with a 95% confidence interval were employed for assess variables. Risk of publication bias was assessed with funnel plots. Results. Data from 7 studies with a total of 2816 AP patients were included. COVID-19 positive was associated with higher incidences of pancreatic necrosis (OR = 1.65; 95% CI: 1.13 to 2.42, = 0.01; = 0.82 for heterogeneity) and persistent organ failure (OR = 6.87; 95% CI: 2.37 to 19.98, = 0.0004; = 0.12 for heterogeneity), especially cardiovascular failure (OR = 2.92; 95% CI: 1.66 to 5.14, = 0.0002; = 0.58 for heterogeneity) and acute respiratory distress syndrome (ARDS) or acute lung injury (ALI) (OR = 3.03; 95% CI: 2.09 to 4.39, < 0.00001; = 0.20 for heterogeneity). COVID-19 infection induced a higher level of CRP (MD = 0.40; 95% CI: 0.16 to 0.64, = 0.001; < 0.00001 for heterogeneity) as well as coagulation disorders involving platelets, prothrombin time, activated partial thromboplastin time, and D-dimer (all < 0.05). During hospitalization, COVID-19 positive was associated with higher ICU admission rate (OR = 2.76; 95% CI: 1.98 to 3.85 < 0.00001; = 0.47 for heterogeneity). COVID-19 positive AP was associated with a higher mortality rate (OR = 3.70; 95% CI: 2.60 to 5.25, < 0.00001; = 0.12 for heterogeneity). Discussion. The number of included studies is limited and none is RCT, thus the risks of publication and selective bias could not be ignored. COVID-19 deteriorated the severity and clinical outcomes of AP, with a high incidence of morbidity and mortality.
- 12 August 2022
- Published by Wolters Kluwer Health in Clinical Nuclear Medicine
- Vol. 47 (12) , 1079-1081
68Ga-FAPI and 18F-FDG PET/CT Images of a Patient With Rosai-Dorfman Disease With Liver Involvement
Rosai-Dorfman disease is a rare non-Langerhans cell histiocytosis that is histopathologically characterized by the accumulation of CD68-positive and S100-positive histiocytes. The pathological changes are mostly discovered in lymph nodes. We report a case of Rosai-Dorfman disease with liver and bone marrow involvement. In this case, 18F-FDG PET/CT revealed FDG avidity in multiple lymph nodes, liver, and bone marrow. 68Ga-FAPI PET/CT showed higher uptake of 68Ga-FAPI than that of 18F-FDG in the same lesions. The findings of 68Ga-FAPI PET/CT in this patient highlighted the potential value of 68Ga-FAPI for staging in Rosai-Dorfman disease.
- 12 August 2022
- Published by Wolters Kluwer Health in Clinical Nuclear Medicine
- Vol. 47 (12) , 1079-1081
68Ga-FAPI and 18F-FDG PET/CT Images of a Patient With Rosai-Dorfman Disease With Liver Involvement
Rosai-Dorfman disease is a rare non-Langerhans cell histiocytosis that is histopathologically characterized by the accumulation of CD68-positive and S100-positive histiocytes. The pathological changes are mostly discovered in lymph nodes. We report a case of Rosai-Dorfman disease with liver and bone marrow involvement. In this case, 18F-FDG PET/CT revealed FDG avidity in multiple lymph nodes, liver, and bone marrow. 68Ga-FAPI PET/CT showed higher uptake of 68Ga-FAPI than that of 18F-FDG in the same lesions. The findings of 68Ga-FAPI PET/CT in this patient highlighted the potential value of 68Ga-FAPI for staging in Rosai-Dorfman disease.
- Open Access
- 17 November 2021
- Published by Wiley in Emergency Medicine International
- Vol. 2021, 1-7
Laparoscopic Intervention to Pancreatic Pseudocyst Confers Short-Term Benefits: A Meta-Analysis
Background. Surgical interventions for pancreatic pseudocyst (PP) are traditionally managed by an open surgical approach. With the development of minimally invasive surgical techniques, a laparoscopic surgical approach for PPs has been conducted increasingly with comparable outcomes. The present study was conducted to compare the efficacy and safety of surgical intervention for PPs between the laparoscopic approach and the open approach. Methods. Databases including Cochrane Library, PubMed, and EMBASE were searched to identify studies that compared the safety and efficacy of surgical intervention for PPs between the laparoscopic approach and the open approach (until Aug 1st 2020). Results. A total of 6 studies were eligible in qualitative synthesis. The laparoscopic approach was associated with less intraoperative blood loss (MD = −69.97; 95% CI: −95.14 to −44.70, ; for heterogeneity) and shorter operating time (MD = −33.12; 95% CI: −62.24 to −4.00, ; for heterogeneity). There was no significant difference found between the two approaches regarding the success rate and the recurrence rate. The postoperative complications and mortality rates were comparable between the two approaches. Conclusions. The laparoscopic approach for the surgical intervention of PPs is safe and efficacious with shorter-term benefits.
- 17 November 2021
- Published by Wiley in Emergency Medicine International
- Vol. 2021, 1-7
Open AccessLaparoscopic Intervention to Pancreatic Pseudocyst Confers Short-Term Benefits: A Meta-Analysis
Background. Surgical interventions for pancreatic pseudocyst (PP) are traditionally managed by an open surgical approach. With the development of minimally invasive surgical techniques, a laparoscopic surgical approach for PPs has been conducted increasingly with comparable outcomes. The present study was conducted to compare the efficacy and safety of surgical intervention for PPs between the laparoscopic approach and the open approach. Methods. Databases including Cochrane Library, PubMed, and EMBASE were searched to identify studies that compared the safety and efficacy of surgical intervention for PPs between the laparoscopic approach and the open approach (until Aug 1st 2020). Results. A total of 6 studies were eligible in qualitative synthesis. The laparoscopic approach was associated with less intraoperative blood loss (MD = −69.97; 95% CI: −95.14 to −44.70, ; for heterogeneity) and shorter operating time (MD = −33.12; 95% CI: −62.24 to −4.00, ; for heterogeneity). There was no significant difference found between the two approaches regarding the success rate and the recurrence rate. The postoperative complications and mortality rates were comparable between the two approaches. Conclusions. The laparoscopic approach for the surgical intervention of PPs is safe and efficacious with shorter-term benefits.