Yuefu Zhan

Dr. Yuefu Zhan

orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

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

Representative Publications

View All Publications
10 June 2023 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
ZXZhenghua Xu
Zhenghua Xu
XZXudong Zhang
Xudong Zhang
HZHexiang Zhang
Hexiang Zhang
YLYunxin Liu
Yunxin Liu
YZYuefu Zhan
Yuefu Zhan
TLThomas Lukasiewicz
Thomas Lukasiewicz

1 December 2024 in Quantitative Imaging in Medicine and Surgery
Verified
Journal Article
Open access
XWXiaocheng Wang
Xiaocheng Wang
LDLingling Deng
Lingling Deng
XLXudong Liu
Xudong Liu
SCShitong Cheng
Shitong Cheng
YZYuefu Zhan
Yuefu Zhan
JCJianqiang Chen
Jianqiang Chen
Cognitive impairment associated with mild-to-moderate chronic traumatic brain injury (TBI) presents substantial challenges for which the functionality of the brain glymphatic system is a key area of interest. This study aimed to explore the functionality of the brain glymphatic system in patients with chronic cognitive impairment following mild-to-moderate TBI using diffusion tensor image analysis along the perivascular space (DTI-ALPS). This was a prospective cross-sectional study. A total of 56 patients with mild-to-moderate chronic TBI at the Radiology Outpatient Clinic of the First Affiliated Hospital of Hainan Medical University were enrolled in the study between January 2021 and July 2022. Additionally, 20 healthy control (HC) subjects were recruited from the Health Screening Center during the same period as the HC group. Relevant clinical data for all the participants were collected, and cognitive assessments were conducted using cognitive scales. The TBI patients were categorized into the traumatic brain injury cognitive impairment (TBI-CI) and traumatic brain injury cognitively normal (TBI-CN) groups based on their of the Chinese Version of the Montreal Cognitive Assessment-Basic (MoCA-BC) scores. Each group comprised 20 subjects. All three groups of participants underwent diffusion tensor imaging (DTI). The DTI data were processed and analyzed using the MRIcron and FMRIB Software Library (FSL) toolboxes, and the analysis along the perivascular space (ALPS) index was calculated. Differences in the ALPS index among the three groups were examined by an analysis of covariance, adjusted for age and gender. A corrected receiver operating characteristic (ROC) curve analysis was employed to evaluate the diagnostic performance of the ALPS index in identifying patients with traumatic chronic cognitive impairment. Compared with the HC group, both the mild-to-moderate TBI patients with and without cognitive impairment had a decreased ALPS index (HC vs. TBI-CI: 1.629 vs. 1.302, P<0.001; HC vs. TBI-CN: 1.629 vs. 1.523, P=0.003). Moreover, the decrease in the ALPS index was more significant in the mild-to-moderate TBI patients with cognitive impairment (TBI-CN vs. TBI-CI: 1.523 vs. 1.302, P<0.001). The adjusted ROC curve analysis revealed that the area under the curve (AUC) of the ALPS index for diagnosing traumatic chronic cognitive impairment was 0.983 [95% confidence interval (CI): 0.953–1, P<0.001], with a sensitivity of 90% and specificity of 95%. Cognitive impairment in patients with mild-to-moderate chronic TBI may be associated with impairment of the glymphatic system. Additionally, the ALPS index may serve as a potential predictor of the disease. Our findings provide some novel insights into the pathophysiological mechanisms underlying cognitive impairment in mild-to-moderate chronic TBI.

21 September 2011 in Experimental and Therapeutic Medicine
Verified
Journal Article
Open access
PDF file
XHXiangjun Han
Xiangjun Han
YZYuefu Zhan
Yuefu Zhan
JCJianqiang Chen
Jianqiang Chen
Evaluation of multi-slice CT angiography (MSCTA) and digital subtraction angiography (DSA) was made through the comparison of the two methods in detecting the blood supply of meningiomas, providing further clues for the improved application of MSCTA and DSA in the blood supply of meningiomas. In this study, 20 patients with meningiomas underwent CT angiography by 16-slice spiral CT and inspection 1 week later using DSA. The blood supply of the meningiomas (including the tumor feeding arteries, draining veins, venous sinuses, vessels around the tumor and skull-shaped images, such as comparative evaluation) was compared. The results showed that MSCTA and DSA inspection clearly depicted tumor feeding arteries, draining veins and venous sinuses in all 20 patients. DSA was slightly more effective than MSCTA in displaying fine branches, and DSA had obvious advantages over MSCTA in displaying vessels near the skull. In the display of the tumor’s compression and pushing over its peripheral vasculature, MSCTA had the three-dimensional advantage, although in displaying parasagittal sinus meningiomas, it may overestimate the presence of venous sinus invasion due to the impact of bone structure. This does not affect DSA. MSCTA accurately assesses the relationship between the tumor and the bone, and provides three-dimensional images for the pre-operative simulation of surgical approaches. Taken together, MSCTA and DSA are both able to provide important information, such as accurate visualization of tumor-feeding arteries, draining veins and venous sinuses, with high consistency. MSCTA is more advantageous in the depiction of three-dimensional relationships between the tumor and peripheral vessels and skull. Compared to DSA, MSCTA is non-invasive, faster and less costly.

12 January 2026 in Frontiers in Bioengineering and Biotechnology
Verified
Journal Article
Open access
PDF file
ZFZiyue Fu
Ziyue Fu
LCLing Chen
Ling Chen
LMLingjiang Mo
Lingjiang Mo
YZYuefu Zhan
Yuefu Zhan
LMLiheng Ma
Liheng Ma
One of the main complications of diabetes is diabetic foot ulcers (DFUs), which have a substantial global impact on morbidity, medical expenses, and lower limb amputations. A complicated microenvironment—characterized by bacterial infections, chronic inflammation, oxidative stress, and poor angiogenesis impedes the healing process of DFUs. Cerium oxide nanoparticles (CeO 2 nanoparticles), owing to their antibacterial, anti-inflammatory, and antioxidant properties, have emerged as a promising treatment. These nanoparticles can scavenge reactive oxygen species (ROS), reduce inflammation, and stimulate tissue regeneration by modulating macrophage polarization and promoting angiogenesis. However, limitations such as poor biocompatibility, aggregation, and dose-dependent toxicity hinder their clinical application. Recent advances in coating strategies—including polymeric, natural biomolecule, and hybrid coatings have improved the stability, targeting, and controlled release of CeO 2 nanoparticles. This review discusses the design, therapeutic mechanisms, and safety considerations of coated CeO 2 nanoparticles in DFU treatment, with emphasis on combination therapies and smart-responsive coatings. Clinical translation challenges and future research directions are also addressed.

1 January 2025 in Quantitative Imaging in Medicine and Surgery
Verified
Journal Article
Open access
PDF file
CLCai-Ying Li
Cai-Ying Li
LCLei Chen
Lei Chen
FMFu-Zhao Ma
Fu-Zhao Ma
JCJian-Qiang Chen
Jian-Qiang Chen
YZYuefu Zhan
Yuefu Zhan
Yi-Xiang WangYi-Xiang Wang
Yi-Xiang Wang
Background: Diffusion-derived ‘vessel density’ (DDVD) is a surrogate of the area of micro-vessels per unit tissue. DDVD is calculated according to: DDVD (b0b50) = Sb0/ROIarea0 − Sb50/ROIarea50, where Sb0 and Sb50 refer to the tissue signal when b is 0 or 50 s/mm2. Due to the complexity of pre-eclampsia (PE), even a combination of risk factors and available tests cannot accurately diagnose or predict PE. This study applies DDVD to assess the perfusion of placenta, and study placenta perfusion disturbance in PE patients. Methods: Diffusion-weighted images with b-values of 0, 50 s/mm2 were acquired in 44 normal pregnancies and 25 patients with PE with a 3.0-T magnet. Diffusion-derived vessel density ratio (DDVDr) was calculated according to: DDVD of placenta/DDVD of fetal brain. Results: The DDVD values of the control placentas [n=44, median: 67.53, 95% confidence interval (CI): 52.84–79.46] were significantly higher than those of the PE patients (n=25, median: 23.66, 95% CI: 17.22–45.57, P<0.0001), while there was no difference of the fetal brain DDVD values of the control cases (median: 29.5, 95% CI: 25.46–34.62) and those of PE patients (median: 33.6, 95% CI: 27.82–35.39, P=0.41). DDVDr results of the control cases (median: 2.00, 95% CI: 1.57–2.50) and PE patients (median: 0.81, 95% CI: 0.53–1.18, P<0.0001) were significantly different, with area under the receiver operating characteristic curve (AUROC) of 0.84 for separation. The AUROC was 0.92 for separation when only cases with gestation age (GA) ≤35 weeks were considered (control n=20, PE n=9). If only PE patients with fetal growth restriction (n=8) were compared with the controls (n=44), then AUROC was 0.96 for the separation. A trend was noted with GA and DDVDr negatively correlated in the control group (r=−0.26, P=0.085); however, such a trend was not observed for PE patients. A trend was observed with systolic blood pressure and DDVDr negatively correlated (r=−0.328, P=0.117). Conclusions: DDVDr may play an important role in patient PE prediction and follow-up.

10 March 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
YLYunxin Liu
Yunxin Liu
DYDi Yuan
Di Yuan
ZXZhenghua Xu
Zhenghua Xu
YZYuefu Zhan
Yuefu Zhan
HZHongwei Zhang
Hongwei Zhang
TLThomas Lukasiewicz
Thomas Lukasiewicz
Existing deep learning methods have achieved significant success in medical image segmentation. However, this success largely relies on stacking advanced modules and architectures, which has created a path dependency. This path dependency is unsustainable, as it leads to increasingly larger model parameters and higher deployment costs. To break this path dependency, we introduce deep reinforcement learning to enhance segmentation performance. However, current deep reinforcement learning methods face challenges such as high training cost, independent iterative processes, and high uncertainty of segmentation masks. Consequently, we propose a Pixel-level Deep Reinforcement Learning model with pixel-by-pixel Mask Generation (PixelDRL-MG) for more accurate and robust medical image segmentation. PixelDRL-MG adopts a dynamic iterative update policy, directly segmenting the regions of interest without requiring user interaction or coarse segmentation masks. We propose a Pixel-level Asynchronous Advantage Actor-Critic (PA3C) strategy to treat each pixel as an agent whose state (foreground or background) is iteratively updated through direct actions. Our experiments on two commonly used medical image segmentation datasets demonstrate that PixelDRL-MG achieves more superior segmentation performances than the state-of-the-art segmentation baselines (especially in boundaries) using significantly fewer model parameters. We also conducted detailed ablation studies to enhance understanding and facilitate practical application. Additionally, PixelDRL-MG performs well in low-resource settings (i.e., 50-shot or 100-shot), making it an ideal choice for real-world scenarios.