You-Zhen Feng

You-Zhen Feng

Auto-generated profile, based on publicly available information

You-Zhen Feng

Department of Radiology, the First Affiliated Hospital, Jinan University, No.613, Huangpu West Road, Tianhe District, Guangzhou 510632, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
11 February 2021 in JMIR Medical Informatics
Verified
Journal Article
Open access
JQJuan Quiroz
Juan Quiroz
YFYou-Zhen Feng
You-Zhen Feng
ZCZhong-Yuan Cheng
Zhong-Yuan Cheng
DRDana Rezazadegan
Dana Rezazadegan
PCPing-Kang Chen
Ping-Kang Chen
XCXiang-Ran Cai
Xiang-Ran Cai
Background: COVID-19 has overwhelmed health systems worldwide. It is important to identify severe cases as early as possible, such that resources can be mobilized and treatment can be escalated. Objective: This study aims to develop a machine learning approach for automated severity assessment of COVID-19 based on clinical and imaging data. Methods: Clinical data—including demographics, signs, symptoms, comorbidities, and blood test results—and chest computed tomography scans of 346 patients from 2 hospitals in the Hubei Province, China, were used to develop machine learning models for automated severity assessment in diagnosed COVID-19 cases. We compared the predictive power of the clinical and imaging data from multiple machine learning models and further explored the use of four oversampling methods to address the imbalanced classification issue. Features with the highest predictive power were identified using the Shapley Additive Explanations framework. Results: Imaging features had the strongest impact on the model output, while a combination of clinical and imaging features yielded the best performance overall. The identified predictive features were consistent with those reported previously. Although oversampling yielded mixed results, it achieved the best model performance in our study. Logistic regression models differentiating between mild and severe cases achieved the best performance for clinical features (area under the curve [AUC] 0.848; sensitivity 0.455; specificity 0.906), imaging features (AUC 0.926; sensitivity 0.818; specificity 0.901), and a combination of clinical and imaging features (AUC 0.950; sensitivity 0.764; specificity 0.919). The synthetic minority oversampling method further improved the performance of the model using combined features (AUC 0.960; sensitivity 0.845; specificity 0.929). Conclusions: Clinical and imaging features can be used for automated severity assessment of COVID-19 and can potentially help triage patients with COVID-19 and prioritize care delivery to those at a higher risk of severe disease.

15 November 2021 in Information
Verified
Journal Article
Open access
PDF file
YFYou-Zhen Feng
You-Zhen Feng
Sidong LiuSidong Liu
Sidong Liu
ZCZhong-Yuan Cheng
Zhong-Yuan Cheng
JQJuan Quiroz
Juan Quiroz
DRDana Rezazadegan
Dana Rezazadegan
XCXiang-Ran Cai
Xiang-Ran Cai
Automatic severity assessment and progression prediction can facilitate admission, triage, and referral of COVID-19 patients. This study aims to explore the potential use of lung lesion features in the management of COVID-19, based on the assumption that lesion features may carry important diagnostic and prognostic information for quantifying infection severity and forecasting disease progression. A novel LesionEncoder framework is proposed to detect lesions in chest CT scans and to encode lesion features for automatic severity assessment and progression prediction. The LesionEncoder framework consists of a U-Net module for detecting lesions and extracting features from individual CT slices, and a recurrent neural network (RNN) module for learning the relationship between feature vectors and collectively classifying the sequence of feature vectors. Chest CT scans of two cohorts of COVID-19 patients from two hospitals in China were used for training and testing the proposed framework. When applied to assessing severity, this framework outperformed baseline methods achieving a sensitivity of 0.818, specificity of 0.952, accuracy of 0.940, and AUC of 0.903. It also outperformed the other tested methods in disease progression prediction with a sensitivity of 0.667, specificity of 0.838, accuracy of 0.829, and AUC of 0.736. The LesionEncoder framework demonstrates a strong potential for clinical application in current COVID-19 management, particularly in automatic severity assessment of COVID-19 patients. This framework also has a potential for other lesion-focused medical image analyses.

26 September 2021 in BMC Nephrology
Verified
Journal Article
Open access
PDF file
YFYou-Zhen Feng
You-Zhen Feng
XCXiao-Qiao Chen
Xiao-Qiao Chen
ZCZhong-Yuan Cheng
Zhong-Yuan Cheng
QLQi-Ting Lin
Qi-Ting Lin
PCPing-Kang Chen
Ping-Kang Chen
XCXiang-Ran Cai
Xiang-Ran Cai
Background: The current study investigated the performance of intravoxel incoherent motion diffusion (IVIM) technology in monitoring early renal injury in streptozotocin rats. Methods: Forty-eight Sprague-Dawley (SD) rats were divided into a control group and a diabetic mellitus (DM) group. Six rats in each group were randomly selected for MR scans at four different time points (0, 4, 8, and 12 weeks). The IVIM-derived parameters (D, D*, f and ADC values) of the renal cortex (CO), outer and inner stripe of the outer medulla (OS, IS), and internal medulla (IM) were acquired. Changes in each IVIM-derived parameter over time were analyzed, and differences between the two groups at each point were assessed. The associations between the IVIM parameters and IV collagen expression, urine volume (UV), blood urea nitrogen (BUN), and serum creatinine (Scr) were investigated. Results: The D and D* values of CO and the ADC values of CO, OS, IS and IM displayed significantly different trends between the two groups over time (P<0.05). In addition, significant correlations were discovered between the D* value of CO and UV and BUN (r=0.527, P=0.033; r=0.617, P=0.005), between the ADC value of IM and BUN (r=0.557, P=0.019) and between the f value of IM and BUN (r=0.527, P=0.033). No correlation was found between IVIM parameters and IV collagen expression and Scr. Conclusions: IVIM is a potential sensitive and noninvasive technology for the simultaneous assessment of early renal cortical and medullary injuries induced by diabetes.

1 November 2016 in Journal of the Neurological Sciences
Verified
Journal Article
Open access
Wenxian LiWenxian Li
Wenxian Li
YFYouzhen Feng
Youzhen Feng
WLWeibiao Lu
Weibiao Lu
XXXie Xie
Xie Xie
ZXZhilin Xiong
Zhilin Xiong
LHLi'an Huang
Li'an Huang
To determine the morphological changes of intracranial arteries and whole-brain perfusion in undetermined isolated vertigo (UIV) patients using 320-detector row computed tomography (CT). A total of 150 patients who underwent CT angiography (CTA) and CT perfusion (CTP) imaging were divided into UIV group and benign paroxysmal positional vertigo (BPPV) group. Sixty individuals with sex- and age-matched without vertigo and cerebral diseases served as the control. The morphological changes of intracranial arteries, perfusion parameters and vascular risk factors (VRFs) were analyzed, calculated and compared. In UIV patients, hypertension (HT), hyperlipidemia and number of VRFs≥3 occurred more commonly (P<0.0125, respectively). The incidence of vertebral artery dominance (VAD), vertebral artery stenosis (VAS) and basilar artery curvature (BAC) were significantly higher (P<0.0125, respectively). HT was an independent risk predictor of non-VAD (OR: 5.411, 95%CI: 1.401; 20.900, P=0.014). HT and VAD associated with BAC served as risk predictors (OR: 4.081, 95%CI: 1.056;15.775, P=0.041 and OR: 6.284, 95%CI: 1.848; 21.365, P=0.003, respectively). The absolute difference in relative values of CTP parameters from cerebellum and brainstem were significantly different (P<0.05), and hypoperfusion was found in the territories of the non-VAD side and the BAC cohort (P<0.05, respectively). On the basis of multiple VRFs, morphological changes of vertebrobasilar artery (VBA) and the unilateral hypoperfusion of the cerebellum and brainstem, that acts as a herald for IV occurrence, which should be paid cautious attention to UIV patients.

9 March 2023 in European Journal of Radiology
Verified
Journal Article
Open access
XYXin Yu
Xin Yu
YHYan-Hao Huang
Yan-Hao Huang
YFYou-Zhen Feng
You-Zhen Feng
ZCZhong-Yuan Cheng
Zhong-Yuan Cheng
CWCunchuan Wang
Cunchuan Wang
XCXiang-Ran Cai
Xiang-Ran Cai
To evaluate predictive values of body composition parameters measured from preoperative CT/MRIs for postoperative complications after laparoscopic sleeve gastrectomy (LSG) and Roux-en-Y gastric bypass (LRYGB) in patients with obesity. In this retrospective case-control study, patients performing abdominal CT/MRIs within one month before and developing 30-day complications after bariatric procedures were matched for age, sex, and type of surgery with patients without complications (1/3 ratio, respectively). Complications were determined by documentation in the medical record. Two readers blindly segmented the total abdominal muscle area (TAMA) and visceral fat area (VFA) using predetermined thresholds for the Hounsfield unit (HU) on unenhanced CT and the signal intensity (SI) on T1-weighted MRI at the L3 vertebral level. Visceral obesity (VO) was defined as VFA > 136 cm2 in males and > 95 cm2 in females. These measures, along with perioperative variables, were compared. Multivariate logistic regression analyses were performed. Of 145 included patients, 36 had postoperative complications. No significant differences between LSG and LRYGB were present regarding complications and VO. Hypertension (p = 0.022), impaired lung function (p = 0.018), American Society of Anesthesiologists (ASA) grade (p = 0.046), VO (p = 0.021), and VFA/TAMA ratio (p < 0.0001) were associated with postoperative complications in the univariate logistic analysis; the VFA/TAMA ratio was the only independent predictor in multivariate analyses (OR 2.01, 95% CI 1.37-2.93, p < 0.001). The VFA/TAMA ratio provides important perioperative information in predicting patients who are likely to develop postoperative complications undergoing bariatric surgery.

13 December 2023 in British Journal of Radiology
Verified
Journal Article
Open access
PDF file
ZCZhong-Yuan Cheng
Zhong-Yuan Cheng
SGShang-Ao Gong
Shang-Ao Gong
PCPing-Kang Chen
Ping-Kang Chen
ZYZong-Chao Yu
Zong-Chao Yu
CQChen Qiu
Chen Qiu
XCXiang-Ran Cai
Xiang-Ran Cai
To validate the feasibility of intravoxel incoherent motion imaging (IVIM) for monitoring renal injury and uric acid-lowering efficacy in a rat model of hyperuricaemia. A total of 92 rats were analysed and categorized into 4 groups: control (CON), hyperuricaemia (HUA), allopurinol intervention (ALL), and combined intervention (COM). Eight rats were randomly selected from each group and underwent IVIM scanning on days 0, 1, 3, 5, 7, and 9. Quantitative magnetic resonance values (D, D*, and f values) measured from the different renal anatomical regions. Quantitative histopathological analysis was performed to assess renal tubular injury using neutrophil gelatinase-associated lipocalin (NGAL), and renal fibrosis using alpha-smooth-muscle-actin (α-SMA). Pearson’s correlation analysis was used to determine the correlation between IVIM-derived parameters and the expression of NGAL and α-SMA. The D values of the HUA, ALL, and COM groups generally showed a downward trend over time, and this fluctuation was most significant in the HUA group. The D values showed significant intergroup differences at each point, whereas only a few discrepancies were found in the D* and f values. In addition, the renal D value was negatively correlated with the positive staining rates for NGAL and α-SMA (P < .05), except for the lack of correlation between Dos and α-SMA (P > .05). IVIM could be a noninvasive and potential assessment modality for the evaluation of renal injury induced by hyperuricaemia and its prognostic efficacy. IVIM could be a surrogate manner in monitoring renal damage induced by hyperuricaemia and its treatment evaluation.