Huiqing Ge

Prof. Dr. Huiqing Ge


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
1 May 2020 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
LZLingwei Zhang
Lingwei Zhang
KMKedong Mao
Kedong Mao
KDKailiang Duan
Kailiang Duan
SFSiqi Fang
Siqi Fang
YLYunfei Lu
Yunfei Lu
QPQing Pan
Qing Pan

30 September 2016 in Respiratory Care
Verified
Journal Article
Open access
PDF file
ZLZhihua Lu
Zhihua Lu
QXQiuping Xu
Qiuping Xu
YYYuehua Yuan
Yuehua Yuan
GZGe Zhang
Ge Zhang
FGFeng Guo
Feng Guo
HGHuiqing Ge
Huiqing Ge
BACKGROUND: Diaphragmatic dysfunction is often underdiagnosed and is among the risk factors for failed weaning. The purpose of this study was to determine the prevalence of diaphragmatic dysfunction diagnosed by B-mode ultrasonography and to determine whether prolonged weaning subjects with diaphragmatic dysfunction have increased duration of mechanical ventilation compared with those without diaphragmatic dysfunction. METHODS: This was a prospective observational study in mechanically ventilated subjects who failed ≥3 spontaneous breathing trials or required >7 d of weaning after the first spontaneous breathing trial. Diaphragm thickness was measured in the zone of apposition using a 6–13-MHz ultrasound transducer during a spontaneous breathing trial. The diaphragmatic thickening fraction was calculated as a percentage from the formula: (Thickness at peak inspiration − thickness at end expiration)/thickness at end expiration. Intra-observer and inter-observer reliability were also evaluated. RESULTS: Forty-one subjects (24 males; 62.2 ± 15.9 y old) were included in the study. Of these, the prevalence of ultrasonographic diaphragmatic dysfunction (defined as diaphragmatic thickening fraction of <20% with inspiration) was 34.1% ( n = 14). Subjects with diaphragmatic dysfunction had longer ventilation time after inclusion (293.4 ± 194.8 vs 145.1 ± 101.3 h, P = .02) and ICU stay (29.2 ± 11.4 vs 22.4 ± 7.7 d, P = .03) than subjects without diaphragmatic dysfunction. CONCLUSIONS: Diaphragmatic dysfunction as assessed by B-mode ultrasonography is common in subjects with prolonged weaning. Subjects with such diaphragmatic dysfunction show longer mechanical ventilation durations and ICU stays.

1 June 2021 in Computer Methods and Programs in Biomedicine
Verified
Journal Article
Closed access
QPQing Pan
Qing Pan
LZLingwei Zhang
Lingwei Zhang
MJMengzhe Jia
Mengzhe Jia
JPJie Pan
Jie Pan
QGQiang Gong
Qiang Gong
LFLuping Fang
Luping Fang

20 October 2018 in BMC Anesthesiology
Verified
Journal Article
Open access
PDF file
HGHuiqing Ge
Huiqing Ge
YJYe Jiang
Ye Jiang
QJQijun Jin
Qijun Jin
LWLinjun Wan
Linjun Wan
XQXiming Qian
Ximing Qian
Zhongheng ZhangZhongheng Zhang
Zhongheng Zhang
Postoperative hypoxemia is quite common in patients with acute aortic dissection (AAD) and is associated with poor clinical outcomes. However, there is no method to predict this potentially life-threatening complication. The study aimed to develop a regression model in patients with AAD to predict postoperative hypoxemia, and to validate it in an independent dataset. All patients diagnosed with AAD from December 2012 to December 2017 were retrospectively screened for potential eligibility. Preoperative and intraoperative variables were included for analysis. Logistic regression model was fit by using purposeful selection procedure. The original dataset was split into training and validating datasets by 4:1 ratio. Discrimination and calibration of the model was assessed in the validating dataset. A nomogram was drawn for clinical utility. A total of 211 patients, involving 168 in non-hypoxemia and 43 in hypoxemia group, were included during the study period (incidence: 20.4%). Duration of mechanical ventilation (MV) was significantly longer in the hypoxemia than non-hypoxemia group (41(10.5140) vs. 12(3.75,70.25) hours; p = 0.002). There was no difference in the hospital mortality rate between the two groups. The purposeful selection procedure identified 8 variables including hematocrit (odds ratio [OR]: 0.89, 95% confidence interval [CI]: 0.80 to 0.98, p = 0.011), PaO2/FiO2 ratio (OR: 0.99, 95% CI: 0.99 to 1.00, p = 0.011), white blood cell count (OR: 1.21, 95% CI: 1.06 to 1.40, p = 0.008), body mass index (OR: 1.32, 95% CI: 1.15 to 1.54; p = 0.000), Stanford type (OR: 0.22, 95% CI: 0.06 to 0.66; p = 0.011), pH (OR: 0.0002, 95% CI: 2*10− 8 to 0.74; p = 0.048), cardiopulmonary bypass time (OR: 0.99, 95% CI: 0.98 to 1.00; p = 0.031) and age (OR: 1.03, 95% CI: 0.99 to 1.08; p = 0.128) to be included in the model. In an independent dataset, the area under curve (AUC) of the prediction model was 0.869 (95% CI: 0.802 to 0.936). The calibration was good by visual inspection. The study developed a model for the prediction of postoperative hypoxemia in patients undergoing operation for AAD. The model showed good discrimination and calibration in an independent dataset that was not used for model training.

17 June 2021 in Sensors
Verified
Journal Article
Open access
PDF file
QPQing Pan
Qing Pan
MJMengzhe Jia
Mengzhe Jia
QLQijie Liu
Qijie Liu
LZLingwei Zhang
Lingwei Zhang
JPJie Pan
Jie Pan
HGHuiqing Ge
Huiqing Ge
Mechanical ventilation is an essential life-support treatment for patients who cannot breathe independently. Patient–ventilator asynchrony (PVA) occurs when ventilatory support does not match the needs of the patient and is associated with a series of adverse clinical outcomes. Deep learning methods have shown a strong discriminative ability for PVA detection, but they require a large number of annotated data for model training, which hampers their application to this task. We developed a transfer learning architecture based on pretrained convolutional neural networks (CNN) and used it for PVA recognition based on small datasets. The one-dimensional signal was converted to a two-dimensional image, and features were extracted by the CNN using pretrained weights for classification. A partial dropping cross-validation technique was developed to evaluate model performance on small datasets. When using large datasets, the performance of the proposed method was similar to that of non-transfer learning methods. However, when the amount of data was reduced to 1%, the accuracy of transfer learning was approximately 90%, whereas the accuracy of the non-transfer learning was less than 80%. The findings suggest that the proposed transfer learning method can obtain satisfactory accuracies for PVA detection when using small datasets. Such a method can promote the application of deep learning to detect more types of PVA under various ventilation modes.

24 November 2023 in BMC Public Health
Verified
Journal Article
Open access
PDF file
YLYuanjiao Liu
Yuanjiao Liu
JZJinghan Zhu
Jinghan Zhu
ZGZiye Guo
Ziye Guo
JYJiazhou Yu
Jiazhou Yu
XZXuhui Zhang
Xuhui Zhang
YZYimin Zhu
Yimin Zhu
Background: Limited evidence is available on the association between estimated cardiorespiratory fitness (e-CRF) and incidence of cardiovascular disease (CVD) in Chinese population. Methods: A total of 10,507 adults including 5084 men (48.4%) and 5423 (51.6%) women with a median age of 56.0 (25% quantile: 49, 75% quantile 63) years from the China Health and Retirement Longitudinal Study (CHARLS) was recruited in 2011 as baseline. The CVD incident events were followed-up until 2018. e-CRF was calculated from sex-specific longitudinal non-exercise equations and further grouped into quartiles. Cox proportional models were used to calculate hazard ratio (HR) and 95% confidence interval (CI) for incidence risks of CVD, heart disease and stroke. Results: During a median follow-up of 7 years, a total of 1862 CVD, 1409 heart disease and 612 stroke events occurred. In fully adjusted models, each one MET increment of e-CRF was associated with lower risk of CVD (HR = 0.91, 95%CI = 0.85–0.96 for males, HR = 0.87, 95%CI = 0.81–0.94 for females). Compared with the Quartile (Q)1 group, the HRs (95%CI) of the Q2, Q3 and Q4 groups were 0.84 (0.63–1.03), 0.72 (0.57–0.91) and 0.66 (0.51–0.87) for CVD in males. Females had HRs of 0.79 (0.66–0.96) in Q2, 0.71 (0.57–0.88) in Q3 and 0.58 (0.45–0.75) in Q4 for CVD. The associations between e-CRF and heart disease and stroke were slightly weaker than that for CVD in both males and females. Conclusions: Higher e-CRF decreases the incident risk of CVD, heart disease and stroke.