Liwei Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Liwei Hu has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Yumin Zhong
Department of Radiology, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China
7 shared publications - Chen Guo
Department of Radiology, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China
4 shared publications - Qian Wang
Department of Radiology, Shanghai Children's Medical Center, Shanghai Jiao Tong University School of Medicine, No.1678 Dong Fang Road, Shanghai 200127, China
4 shared publications - Rongzhen Ouyang
Department of Radiology, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University, Shanghai 200120, China
3 shared publications - XiaoFen Yao
Department of Radiology, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China
3 shared publications - Aimin Sun
Department of Radiology, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University, Shanghai 200120, China
2 shared publications
Latest Publications
- 14 January 2026
- Published by Springer Nature in Chinese Journal of Academic Radiology
- Vol. 9 (1) , 49-56
The study of altered brain structure in preterm infants with low grade intraventricular hemorrhage utilizing 3D T1WI whole brain MR images
- 14 January 2026
- Published by Springer Nature in Chinese Journal of Academic Radiology
- Vol. 9 (1) , 49-56
The study of altered brain structure in preterm infants with low grade intraventricular hemorrhage utilizing 3D T1WI whole brain MR images
- 18 October 2024
- Published by Oxford University Press (OUP) in Radiation Protection Dosimetry
- Vol. 200 (20) , 1981-1989
From age-specific to size-specific dose protocol for paediatric head computed tomography: a simple practical strategy for necessity assessment and parameter setting
This study proposes a practical approach to assessing the need for and determining the transitioning parameters from age-specific to size-specific dose protocols in paediatric head computed tomography examinations. It was designed to begin with the two different age-specific protocols currently applied on paediatric patients of 1 y ≤ age ≤ 6 y (21.34 mGy) and those of age >6 y (34.73 mGy). One characteristic size (HDthreshold) was found by measuring anteroposterior head diameter on historical topograms and used with age to set 135 cases into four groups for investigation (A, ≤6y, ≤HDthreshold; B, ≤6y, >HDthreshold; C, >6y, ≤HDthreshold; and D, >6 y, >HDthreshold). The image quality was compared in both subjective and objective manners. Results suggest patients in Group C may have received a potential overdose (38%), while the exposure setting for Group D may replace that for B. The proposed strategy was sufficient to assess the necessity and determine the transitioning parameters from age- to size-specific dose protocols.
- 18 October 2024
- Published by Oxford University Press (OUP) in Radiation Protection Dosimetry
- Vol. 200 (20) , 1981-1989
From age-specific to size-specific dose protocol for paediatric head computed tomography: a simple practical strategy for necessity assessment and parameter setting
This study proposes a practical approach to assessing the need for and determining the transitioning parameters from age-specific to size-specific dose protocols in paediatric head computed tomography examinations. It was designed to begin with the two different age-specific protocols currently applied on paediatric patients of 1 y ≤ age ≤ 6 y (21.34 mGy) and those of age >6 y (34.73 mGy). One characteristic size (HDthreshold) was found by measuring anteroposterior head diameter on historical topograms and used with age to set 135 cases into four groups for investigation (A, ≤6y, ≤HDthreshold; B, ≤6y, >HDthreshold; C, >6y, ≤HDthreshold; and D, >6 y, >HDthreshold). The image quality was compared in both subjective and objective manners. Results suggest patients in Group C may have received a potential overdose (38%), while the exposure setting for Group D may replace that for B. The proposed strategy was sufficient to assess the necessity and determine the transitioning parameters from age- to size-specific dose protocols.
- Open Access
- 6 September 2023
- Published by MDPI AG in Preprints.org
Hemodynamic Disturbances Assess Cardiac Inefficiency in Repaired Fontan Patients: A 4D Flow Cardiovascular Magnetic Resonance Study
Objectives: The purposes of this study were to 1) analyze the disruption of intracardiac flow by flow components and kinetic energy (KE) parameters in single ventricle physiology patients and compare with controls; 2) investigate left ventricular diastolic dysfunction in terms of 4D flow parameters in repaired Fontan (rFontan) group; 3) compare intracardiac flow parameters in morphologic LV and RV rFontan patient groups. Methods: Twenty-five rFontan patients (age: 10±3, M/F: 15/10) and fourteen controls (age: 10±2, M/F: 8/6) were prospectively recruited. All underwent cine and 4D flow cardiovascular magnetic resonance (CMR) on 3.0T scanner. Cardiac function and inter-ventricular mechanical dyssynchrony were analyzed using cine images. From 4D flow CMR, ventricular flow components were assessed: direct flow, retained inflow, delayed ejection flow and residual volume. Global and regional blood flow KE parameters, normalized to end-diastolic volume (EDV) were analyzed for functional single ventricle (FSV). Left ventricular (LV) diastolic dysfunction was assessed on echocardiography guidelines. Results: In comparison of rFontan vs. controls, median FSV residual volume (28% vs. 23%, P= 0.034) were higher in rFontan; median FSV direct flow (32% vs. 40%, P=0.005) and delayed ejection flow (17% vs. 24%, P=0.024) were lower in rFontan. FSV KEiEDV parameters were all lower in rFontan (all P<0.05). No significant differences were observed for flow components and KE parameters between patients with and without inter-ventricular dyssynchrony. FSV direct flow (AUC=0.76, Sensitivity=86%, Specificity=70%) was an independent predictor of LV diastolic dysfunction. Residual volume and E-wave KEiEDV were significantly different between morphologic RV and morphologic LV patient group. Conclusions: The changed flow pattern and decreased KE may be hemodynamic disturbances and impaired ventricular filling in rFontan patients. Reduced direct flow is associated with LV diastolic dysfunction. Morphologic right ventricular subgroup is worse than morphologic left ventricular group in terms of the intracardiac hemodynamics.
- 6 September 2023
- Published by MDPI AG in Preprints.org
Open AccessHemodynamic Disturbances Assess Cardiac Inefficiency in Repaired Fontan Patients: A 4D Flow Cardiovascular Magnetic Resonance Study
Objectives: The purposes of this study were to 1) analyze the disruption of intracardiac flow by flow components and kinetic energy (KE) parameters in single ventricle physiology patients and compare with controls; 2) investigate left ventricular diastolic dysfunction in terms of 4D flow parameters in repaired Fontan (rFontan) group; 3) compare intracardiac flow parameters in morphologic LV and RV rFontan patient groups. Methods: Twenty-five rFontan patients (age: 10±3, M/F: 15/10) and fourteen controls (age: 10±2, M/F: 8/6) were prospectively recruited. All underwent cine and 4D flow cardiovascular magnetic resonance (CMR) on 3.0T scanner. Cardiac function and inter-ventricular mechanical dyssynchrony were analyzed using cine images. From 4D flow CMR, ventricular flow components were assessed: direct flow, retained inflow, delayed ejection flow and residual volume. Global and regional blood flow KE parameters, normalized to end-diastolic volume (EDV) were analyzed for functional single ventricle (FSV). Left ventricular (LV) diastolic dysfunction was assessed on echocardiography guidelines. Results: In comparison of rFontan vs. controls, median FSV residual volume (28% vs. 23%, P= 0.034) were higher in rFontan; median FSV direct flow (32% vs. 40%, P=0.005) and delayed ejection flow (17% vs. 24%, P=0.024) were lower in rFontan. FSV KEiEDV parameters were all lower in rFontan (all P<0.05). No significant differences were observed for flow components and KE parameters between patients with and without inter-ventricular dyssynchrony. FSV direct flow (AUC=0.76, Sensitivity=86%, Specificity=70%) was an independent predictor of LV diastolic dysfunction. Residual volume and E-wave KEiEDV were significantly different between morphologic RV and morphologic LV patient group. Conclusions: The changed flow pattern and decreased KE may be hemodynamic disturbances and impaired ventricular filling in rFontan patients. Reduced direct flow is associated with LV diastolic dysfunction. Morphologic right ventricular subgroup is worse than morphologic left ventricular group in terms of the intracardiac hemodynamics.
- Open Access
- 31 October 2021
- Published by Springer Nature in BMC Medical Imaging
- Vol. 21 (1) , 1-11
Right and left ventricular function and flow quantification in pediatric patients with repaired tetralogy of Fallot using four-dimensional flow magnetic resonance imaging
To assess the accuracy and reproducibility of right ventricular (RV) and left ventricular (LV) function and flow measurements in children with repaired tetralogy of Fallot (rTOF) using four-dimensional (4D) flow, compared with conventional two-dimensional (2D) magnetic resonance imaging (MRI) sequences. Thirty pediatric patients with rTOF were retrospectively enrolled to undergo 2D balanced steady-state free precession cine (2D b-SSFP cine), 2D phase contrast (PC), and 4D flow cardiac MRI. LV and RV volumes and flow in the ascending aorta (AAO) and main pulmonary artery (MPA) were quantified. Pearson’s or Spearman’s correlation tests, paired t-tests, the Wilcoxon signed-rank test, Bland–Altman analysis, and intraclass correlation coefficients (ICC) were performed. The 4D flow scan time was shorter compared with 2D sequences (P < 0.001). The biventricular volumes between 4D flow and 2D b-SSFP cine had no significant differences (P > 0.05), and showed strong correlations (r > 0.90, P < 0.001) and good consistency. The flow measurements of the AAO and MPA between 4D flow and 2D PC showed moderate to good correlations (r > 0.60, P < 0.001). There was good internal consistency in cardiac output. There was good intraobserver and interobserver biventricular function agreement (ICC > 0.85). RV and LV function and flow quantification in pediatric patients with rTOF using 4D flow MRI can be measured accurately and reproducibly compared to those with conventional 2D sequences.
- 31 October 2021
- Published by Springer Nature in BMC Medical Imaging
- Vol. 21 (1) , 1-11
Open AccessRight and left ventricular function and flow quantification in pediatric patients with repaired tetralogy of Fallot using four-dimensional flow magnetic resonance imaging
To assess the accuracy and reproducibility of right ventricular (RV) and left ventricular (LV) function and flow measurements in children with repaired tetralogy of Fallot (rTOF) using four-dimensional (4D) flow, compared with conventional two-dimensional (2D) magnetic resonance imaging (MRI) sequences. Thirty pediatric patients with rTOF were retrospectively enrolled to undergo 2D balanced steady-state free precession cine (2D b-SSFP cine), 2D phase contrast (PC), and 4D flow cardiac MRI. LV and RV volumes and flow in the ascending aorta (AAO) and main pulmonary artery (MPA) were quantified. Pearson’s or Spearman’s correlation tests, paired t-tests, the Wilcoxon signed-rank test, Bland–Altman analysis, and intraclass correlation coefficients (ICC) were performed. The 4D flow scan time was shorter compared with 2D sequences (P < 0.001). The biventricular volumes between 4D flow and 2D b-SSFP cine had no significant differences (P > 0.05), and showed strong correlations (r > 0.90, P < 0.001) and good consistency. The flow measurements of the AAO and MPA between 4D flow and 2D PC showed moderate to good correlations (r > 0.60, P < 0.001). There was good internal consistency in cardiac output. There was good intraobserver and interobserver biventricular function agreement (ICC > 0.85). RV and LV function and flow quantification in pediatric patients with rTOF using 4D flow MRI can be measured accurately and reproducibly compared to those with conventional 2D sequences.
- 8 June 2021
- Published by Springer Nature in The International Journal of Cardiovascular Imaging
- Vol. 37 (11) , 3255-3267
Impairment of left atrial function in pediatric patients with repaired tetralogy of Fallot: a cardiovascular magnetic resonance imaging study
We aimed to assess left atrial (LA) strain before LA dilatation in patients with repaired tetralogy of Fallot (rTOF) compared with healthy controls. We also determined the effects of right atrial (RA) dilatation on LA performance using cardiovascular magnetic resonance-feature tracking (CMR-FT). Forty-nine pediatric patients with rTOF and 36 age- and sex-matched healthy controls were prospectively recruited between June 2017 and August 2019. Balanced steady-state free precession (2D b-SSFP) cine, 2D late gadolinium enhancement (LGE) and phase-contrast (PC) sequences were acquired on 1.5 and 3.0 Tesla scanners. Both ventricular and atrial volumes and ejection fraction were measured. Left ventricular (LV) strain and diastolic strain rates were evaluated between the rTOF patient and control groups. LA reservoir (Ɛs), conduit (Ɛe), and booster strain (Ɛa) were determined at LV end-systole, LV diastasis, and pre-LA systole, respectively. The first derivatives of the respective strains yielded corresponding peak strain rates. Statistical analysis was performed using the t-test and Mann–Whitney test for parametric and non-parametric variables, respectively. Correlations were assessed using Pearson’s correlation coefficient for normally distributed variables and Spearman’s correlation coefficient for non-parametric data. Intra-observer and inter-observer variabilities of LA strain and strain rate measurements were determined from ten randomly selected rTOF patients and ten control subjects. LA strain was significantly lower in patients with rTOF compared with controls (Ɛs, P < 0.001; Ɛe, P = 0.002; Ɛa, P < 0.001). The correlations between LA strain and RA stroke volume indices (SVi) and RA ejection fraction (EF) were moderate (Ɛs and SVi, r = 0.538, P < 0.001; Ɛs and RA EF, r = 0.493, P < 0.001; Ɛe and SVi, r = 0.532, P < 0.001; Ɛe and RA EF, r = 0.466, P < 0.001). LA strain and strain rates had good reproducibility in intra-observer and inter-observer analyses. LA strain and strain rates decreased in pediatric patients with rTOF compared with controls before LA enlargement. A dysfunction in LA performance might precede LV dysfunction in patients with rTOF, even in the early stages after repair.
- 8 June 2021
- Published by Springer Nature in The International Journal of Cardiovascular Imaging
- Vol. 37 (11) , 3255-3267
Impairment of left atrial function in pediatric patients with repaired tetralogy of Fallot: a cardiovascular magnetic resonance imaging study
We aimed to assess left atrial (LA) strain before LA dilatation in patients with repaired tetralogy of Fallot (rTOF) compared with healthy controls. We also determined the effects of right atrial (RA) dilatation on LA performance using cardiovascular magnetic resonance-feature tracking (CMR-FT). Forty-nine pediatric patients with rTOF and 36 age- and sex-matched healthy controls were prospectively recruited between June 2017 and August 2019. Balanced steady-state free precession (2D b-SSFP) cine, 2D late gadolinium enhancement (LGE) and phase-contrast (PC) sequences were acquired on 1.5 and 3.0 Tesla scanners. Both ventricular and atrial volumes and ejection fraction were measured. Left ventricular (LV) strain and diastolic strain rates were evaluated between the rTOF patient and control groups. LA reservoir (Ɛs), conduit (Ɛe), and booster strain (Ɛa) were determined at LV end-systole, LV diastasis, and pre-LA systole, respectively. The first derivatives of the respective strains yielded corresponding peak strain rates. Statistical analysis was performed using the t-test and Mann–Whitney test for parametric and non-parametric variables, respectively. Correlations were assessed using Pearson’s correlation coefficient for normally distributed variables and Spearman’s correlation coefficient for non-parametric data. Intra-observer and inter-observer variabilities of LA strain and strain rate measurements were determined from ten randomly selected rTOF patients and ten control subjects. LA strain was significantly lower in patients with rTOF compared with controls (Ɛs, P < 0.001; Ɛe, P = 0.002; Ɛa, P < 0.001). The correlations between LA strain and RA stroke volume indices (SVi) and RA ejection fraction (EF) were moderate (Ɛs and SVi, r = 0.538, P < 0.001; Ɛs and RA EF, r = 0.493, P < 0.001; Ɛe and SVi, r = 0.532, P < 0.001; Ɛe and RA EF, r = 0.466, P < 0.001). LA strain and strain rates had good reproducibility in intra-observer and inter-observer analyses. LA strain and strain rates decreased in pediatric patients with rTOF compared with controls before LA enlargement. A dysfunction in LA performance might precede LV dysfunction in patients with rTOF, even in the early stages after repair.
- 6 January 2017
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 90 (1070) , 20160669
A prospective evaluation of contrast and radiation dose and image quality in cardiac CT in children with complex congenital heart disease using low-concentration iodinated contrast agent and low tube voltage and current
Objective: To the assess image quality, contrast dose and radiation dose in cardiac CT in children with congenital heart disease (CHD) using low-concentration iodinated contrast agent and low tube voltage and current in comparison with standard dose protocol. Methods: 110 patients with CHD were randomized to 1 of the 2 scan protocols: Group A (n = 45) with 120 mA tube current and contrast agent of 270 mgI/ml in concentration (Visipaque™; GE Healthcare Ireland, Co., Cork, UK); and Group B (n = 65) with the conventional 160 mA and 370 mgI/ml concentration contrast (Iopamiro®; Shanghai Bracco Sine Pharmaceutical Corp Ltd, Shanghai, China). Both groups used 80 kVp tube voltage and were reconstructed with 70% adaptive statistical iterative reconstruction algorithm. The CT value and noise in aortic arch were measured and the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. A five-point scale was used to subjectively evaluate image quality. Contrast and radiation dose were recorded. Results: There was no difference in age and weight between the two groups (all p > 0.05). The iodine load and radiation dose in Group A were statistically lower (3976 ± 747 mgI vs 5763 ± 1018 mgI in iodine load and 0.60 ± 0.08 mSv vs 0.77 ± 0.10 mSv in effective dose; p < 0.001). However, image noise, CT value, CNR, SNR and subjective image quality for the two groups were similar (all p > 0.05), and with good agreement between the two observers. Comparing the surgery results, the diagnostic accuracy for extracardiac and intracardiac defects for Group A was 96% and 92%, respectively, while the corresponding numbers for Group B were 95% and 93%. Conclusion: Compared with the standard dose protocol, the use of low tube voltage (80 kVp), low tube current (120 mA) and low-concentration iodinated contrast agent (270 mgI/ml) enables a reduction of 30% in iodine load and 22% in radiation dose while maintaining compatible image quality and diagnostic accuracy. Advances in knowledge: The new cardiac CT scanning protocol can largely reduce the adverse effects of radiation and contrast media to children. Meanwhile, it also can be used effectively to examine complex CHD.
- 6 January 2017
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 90 (1070) , 20160669
A prospective evaluation of contrast and radiation dose and image quality in cardiac CT in children with complex congenital heart disease using low-concentration iodinated contrast agent and low tube voltage and current
Objective: To the assess image quality, contrast dose and radiation dose in cardiac CT in children with congenital heart disease (CHD) using low-concentration iodinated contrast agent and low tube voltage and current in comparison with standard dose protocol. Methods: 110 patients with CHD were randomized to 1 of the 2 scan protocols: Group A (n = 45) with 120 mA tube current and contrast agent of 270 mgI/ml in concentration (Visipaque™; GE Healthcare Ireland, Co., Cork, UK); and Group B (n = 65) with the conventional 160 mA and 370 mgI/ml concentration contrast (Iopamiro®; Shanghai Bracco Sine Pharmaceutical Corp Ltd, Shanghai, China). Both groups used 80 kVp tube voltage and were reconstructed with 70% adaptive statistical iterative reconstruction algorithm. The CT value and noise in aortic arch were measured and the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. A five-point scale was used to subjectively evaluate image quality. Contrast and radiation dose were recorded. Results: There was no difference in age and weight between the two groups (all p > 0.05). The iodine load and radiation dose in Group A were statistically lower (3976 ± 747 mgI vs 5763 ± 1018 mgI in iodine load and 0.60 ± 0.08 mSv vs 0.77 ± 0.10 mSv in effective dose; p < 0.001). However, image noise, CT value, CNR, SNR and subjective image quality for the two groups were similar (all p > 0.05), and with good agreement between the two observers. Comparing the surgery results, the diagnostic accuracy for extracardiac and intracardiac defects for Group A was 96% and 92%, respectively, while the corresponding numbers for Group B were 95% and 93%. Conclusion: Compared with the standard dose protocol, the use of low tube voltage (80 kVp), low tube current (120 mA) and low-concentration iodinated contrast agent (270 mgI/ml) enables a reduction of 30% in iodine load and 22% in radiation dose while maintaining compatible image quality and diagnostic accuracy. Advances in knowledge: The new cardiac CT scanning protocol can largely reduce the adverse effects of radiation and contrast media to children. Meanwhile, it also can be used effectively to examine complex CHD.