MZMengsu Zeng

Mengsu Zeng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mengsu Zeng has a total of 34 co-authors. Here are the top co-authors by number of shared publications:

  • Yunfei Zhang

    Shanghai Institute of Medical Imaging Fudan University Shanghai China

    13 shared publications
  • Chun Yang

    Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai, China

    11 shared publications
  • Ruofan Sheng

    Department of Radiology, Zhongshan Hospital Fudan University Shanghai China

    11 shared publications
  • Yongming Dai

    School of Biomedical Engineering ShanghaiTech University Shanghai China

    6 shared publications
  • Xianling Qian

    Department of Radiology, Zhongshan Hospital, Fudan University, Shanghai, China

    4 shared publications
  • Yongming Dai

    School of Biomedical Engineering, State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech Univerisity, Shanghai, China

    3 shared publications

Latest Publications

  • Open Access

    Preoperative prediction of HCC microvascular invasion by multimodel peritumoral diffusion MRI

    by Yunfei Zhang, Jiejun Chen, Xianling Qian, Mengsu Zeng

    Objective: Most studies emphasize intratumoral diffusion-weighted imaging (DWI) metrics, but few studies have examined whether peritumoral DWI metrics can provide added value. We aimed to evaluate three-dimensional, multiscale, multimodel DWI parameters within intratumoral and peritumoral volumes of interest for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC). Materials and methods: Seventy patients with HCC, including 42 with MVI, were prospectively enrolled. We calculated parameters from monoexponential, intravoxel incoherent motion, diffusion kurtosis imaging (DKI), stretched-exponential model (SEM), and continuous-time random-walk models within the tumor and in 3-mm (P3) and 5-mm (P5) peritumoral rims. Features showing differences on univariable analysis were further screened using logistic regression with variance inflation factor filtering to build combined models. Diagnostic performance was assessed by areas under the receiver operating characteristic curve (AUROC). Results: Four intratumoral diffusion-coefficient–related metrics showed significant differences between groups. In the peritumoral regions, 12 parameters differed between groups, including diffusion-coefficient–related indices and heterogeneity-related metrics. Among single parameters, P3_DKI_ K app achieved the best discrimination for MVI (AUROC = 0.713); P5_DKI_ K app and Tumor_SEM_DDC were selected as independent imaging predictors. After integrating alpha-fetoprotein (AFP) and tumor size, the combined model (Cli-IntraPeriDWI) yielded the best performance (AUROC = 0.844). SHAP analysis indicated that higher AFP, larger tumor size, higher P5_DKI_ K app , and lower Tumor_SEM_DDC were associated with increased MVI risk. Conclusion: Peritumoral DWI offers multiple biomarkers and outperforms intratumoral metrics for MVI prediction. An interpretable model integrating peritumoral and intratumoral DWI with clinical factors enables accurate, noninvasive preoperative MVI risk stratification in HCC. Key Points: Question Can multiscale, multimodel peritumoral DWI metrics improve preoperative prediction of MVI in HCC compared with intratumoral DWI metrics? Findings Compared with intratumoral DWI, peritumoral DWI offered more potential imaging biomarkers and superior diagnostic capability for the preoperative diagnosis of MVI in HCC. Relevance statement By integrating intratumoral and peritumoral metrics with clinical risk factors, we developed a model achieving optimal performance and transparent decision interpretability for MVI assessment, supporting preoperative prognostic evaluation and individualized treatment planning in HCC.

  • Development and Validation of a Deep Learning–enabled Single Breath-hold Abbreviated MRI Protocol for Hepatocellular Carcinoma Diagnosis

    by Yunfei Zhang, Zhijun Geng, Xianling Qian, Yuyao Xiao, Ruofan Sheng, Mengsu Zeng

    A deep learning-based single breath-hold abbreviated MRI protocol for hepatocellular carcinoma diagnosis was developed, enabling gadolinium-free, ultrafast imaging while preserving full-sequence diagnostic information and demonstrating performance comparable to that of conventional complete MRI.


  • Virtual Multi‐Phase Contrast Enhanced Liver MRI Using Deep Learning for Evaluating Hepatocellular Carcinoma

    by Yunfei Zhang, Xianling Qian, Changwu Zhou, Fei Wu, Ruofan Sheng, Yongming Dai, Mengsu Zeng

    A deep learning (DL) model was developed to generate contrast‐enhanced MRI ( CE ‐MRI) at multiple enhancement phases (arterial, portal venous, transitional, and hepatobiliary phases) for detecting hepatocellular carcinoma (HCC). In total, 717 patients with HCC or other non‐HCC liver diseases were included, and the DL model was trained to synthesize CE ‐MRI. Three radiologists were invited to assess image quality, diagnostic performance, utility for the Liver Imaging Reporting and Data System (LI‐RADS), and image artifacts. The image quality of DL‐synthesized CE ‐MRI was non‐inferior to that of actual CE ‐MRI ( p ‐values < 0.001), and there was excellent agreement between DL‐synthesized and actual CE ‐MRI for visualizing the LI‐RADS major features. The diagnostic performance of DL‐synthesized CE ‐MRI for HCC (sensitivity = 0.880, specificity = 0.950, AUC = 0.915) was non‐inferior to that of actual CE ‐MRI. Notably, the proposed DL model required only 0.20–0.60 s to obtain multi‐phase CE ‐MRI, compared with more than 20 min needed for actual multi‐phase CE ‐MRI. Overall, the observed advantages of the proposed DL‐based strategy included eliminating the need for GBCAs, high time efficiency, comparable image quality, high diagnostic performance, validity for LI‐RADS, and robustness to image artifacts, indicating potential for clinical translation that may benefit both patients and healthcare providers.

  • Open Access

    MRI-based Intra- and Peritumoral Heterogeneity in Hepatocellular Carcinoma for Microvascular Invasion Prediction and Prognostic Risk Stratification

    by Yunfei Zhang, Shutong Wang, Mingyue Song, Ruofan Sheng, Zhijun Geng, Weiguo Zhang, Mengsu Zeng

    An MRI-based approach for quantifying intra- and peritumoral heterogeneity in hepatocellular carcinoma was developed, showing potential for accurately identifying microvascular invasion and stratifying prognostic risk.


  • Open Access

    Deep learning empowered gadolinium-free contrast-enhanced abbreviated MRI for diagnosing hepatocellular carcinoma

    by Yunfei Zhang, Ruofan Sheng, Xianling Qian, Heqing Wang, Fei Wu, Haoran Dai, Mingyue Song, Chun Yang, Jianjun Zhou, Weiguo Zhang, et al.


  • Detecting microvascular invasion in hepatocellular carcinoma using the impeded diffusion fraction technique to sense macromolecular coordinated water

    by Yunfei Zhang, Ruofan Sheng, Chun Yang, Yongming Dai, Mengsu Zeng

    Objectives Impeded diffusion fraction (IDF) is a novel and promising diffusion-weighted imaging (DWI) technique that allows for the detection of various diffusion compartments, including macromolecular coordinated water, free diffusion, perfusion, and cellular free water. This study aims to investigate the clinical potential of IDF-DWI in detecting microvascular invasion (MVI) in hepatocellular carcinoma (HCC). Methods 66 patients were prospectively included. Metrics derived from IDF-DWI and the apparent diffusion coefficient (ADC) were calculated. Multivariate logistic regression was employed to identify clinical risk factors. Diagnostic performance was evaluated using the area under the receiver operating characteristics curve (AUC-ROC), the area under the precision-recall curve (AUC-PR), and the calibration error (cal-error). Additionally, a power analysis was conducted to determine the required sample size. Results The results suggested a significantly higher fraction of impeded diffusion (FID) originating from IDF-DWI in MVI-positive HCCs (p < 0.001). Moreover, the ADC was found to be significantly lower in MVI-positive HCCs (p = 0.019). Independent risk factors of MVI included larger tumor size and elevated alpha-fetoprotein (AFP) levels. The nomogram model incorporating ADC, FID, tumor size, and AFP level yielded the highest diagnostic accuracy for MVI (AUC-PR = 0.804, AUC-ROC = 0.783, cal-error = 0.044), followed by FID (AUC-PR = 0.693, AUC-ROC = 0.760, cal-error = 0.060) and ADC (AUC-PR = 0.570, AUC-ROC = 0.651, cal-error = 0.164). Conclusion IDF-DWI shows great potential in noninvasively, accurately, and preoperatively detecting MVI in HCC and may offer clinical benefits for prognostic prediction and determination of treatment strategy. Graphical abstract


  • The value of varying diffusion curvature MRI for assessing the microvascular invasion of hepatocellular carcinoma

    by Yunfei Zhang, Ruofan Sheng, Yongming Dai, Chun Yang, Mengsu Zeng

    Purpose Varying diffusion curvature (VDC) MRI is an emerging diffusion-weighted imaging (DWI) technique that can capture non-Gaussian diffusion behavior and reflect tissue heterogeneity. However, its clinical utility has hardly been evaluated. We aimed to investigate the value of the VDC technique in noninvasively assessing microvascular invasion (MVI) in hepatocellular carcinoma (HCC). Methods 74 patients with HCCs, including 39 MVI-positive and 35 MVI-negative HCCs were included into this prospective study. Quantitative metrics between subgroups, clinical risk factors, as well as diagnostic performance were evaluated. The power analysis was also carried out to determine the statistical power. Results MVI-positive HCCs exhibited significantly higher VDC-derived structural heterogeneity measure, D1 (0.680 ± 0.100 × 10–3 vs 0.572 ± 0.148 × 10–3 mm2/s, p = 0.001) and lower apparent diffusion coefficient (ADC) (1.350 ± 0.166 × 10–3 vs 1.471 ± 0.322 × 10–3 mm2/s, p = 0.0495) compared to MVI-negative HCCs. No statistical significance was observed for VDC-derived diffusion coefficient, D0 between the subgroups (p = 0.562). Tumor size (odds ratio (OR) = 1.242) and alpha-fetoprotein (AFP) (OR = 2.527) were identified as risk factors for MVI. A predictive nomogram was constructed based on D1, ADC, tumor size, and AFP, which exhibited the highest diagnostic accuracy (AUC = 0.817), followed by D1 (AUC = 0.753) and ADC (AUC = 0.647). The diagnostic performance of the nomogram-based model was also validated by the calibration curve and decision curve. Conclusion VDC can aid in the noninvasive and preoperative diagnosis of HCC with MVI, which may result in the clinical benefit in terms of prognostic prediction and clinical decision-making. Graphical abstract


  • Evaluate the Microvascular Invasion of Hepatocellular Carcinoma (≤5 cm) and Recurrence Free Survival with Gadoxetate Disodium‐Enhanced MRI‐Based Habitat Imaging

    by Yunfei Zhang, Chun Yang, Xianling Qian, Yongming Dai, Mengsu Zeng

    Background: Tumors are heterogenous and consist of subregions, also known as tumoral habitats, each exhibiting varied biological characteristics. Each habitat corresponds to a cluster of tissue sharing similar structural, metabolic, or functional characteristics. The habitat imaging technique facilitates both the visualization and quantification of these tumoral habitats.Purpose: To evaluate the microvascular invasion (MVI) in hepatocellular carcinoma (HCC) (≤5 cm) and assess the recurrence‐free survival (RFS) using gadoxetate disodium‐enhanced MRI‐based habitat imaging.Study Type: Retrospective.Subjects: 180 patients (52.9 years ± 11.7, 156 men) with HCC.Field Strength/Sequence: 1.5T/contrast‐enhanced T1‐weighted gradient‐echo sequence.Assessment: The enhancement ratio of signal intensity at the arterial phase (AER) and hepatobiliary phase (HBPER) were calculated. The HCC lesions and their peritumoral tissues of 3, 5, and 7 mm were encoded into four habitats. The volume fraction of each habitat was then quantified. The diagnostic performance was assessed using the receiver operating characteristic analysis with 5‐fold cross‐validation. The RFS was evaluated with Kaplan–Meier curves.Results: Habitat 2 (with median to high AER and low HBPER) within the peritumoral tissue of 3 mm (f2‐P3) and tumor diameter could serve as independent risk factors for MVI and showed the statistical significance (odds ratio (OR) of f2‐P3 = 1.170, 95% CI = 1.099–1.246; OR of tumor diameter: 6.112, 95% CI = 2.162–17.280). A nomogram was developed by incorporating f2‐P3 and tumor diameter, demonstrating high diagnostic accuracy. The area under the curve from 5‐fold cross‐validation ranged from 0.880 to 1.000. Additionally, the nomogram model demonstrated high efficacy in risk stratification for RFS.Conclusion: Habitat imaging of HCC and its peritumoral microenvironment has the potential for noninvasive and preoperative identification of MVI and prognostic assessment.Level of Evidence: 3Technical Efficacy: Stage 2


  • Open Access

    The Feasibility of Using Tri-Exponential Intra-Voxel Incoherent Motion DWI for Identifying the Microvascular Invasion in Hepatocellular Carcinoma [Corrigendum]

    by Yunfei Zhang, Ruofan Sheng, Chun Yang, Yongming Dai, Mengsu Zeng

    Corrigendum for the article The Feasibility of Using Tri-Exponential Intra-Voxel Incoherent Motion DWI for Identifying the Microvascular Invasion in Hepatocellular Carcinoma

  • Preoperative prediction of microvascular invasion in hepatocellular carcinoma using diffusion-weighted imaging–based habitat imaging

    by Yunfei Zhang, Jiejun Chen, Chun Yang, Yongming Dai, Mengsu Zeng

    S Objectives Habitat imaging allows for the quantification and visualization of various subregions within the tumor. We aim to develop an approach using diffusion-weighted imaging (DWI)–based habitat imaging for preoperatively predicting the microvascular invasion (MVI) of hepatocellular carcinoma (HCC). Methods Sixty-five patients were prospectively included and underwent multi-b DWI examinations. Based on the true diffusion coefficient (Dt), perfusion fraction (f), and mean kurtosis coefficient (MK), which respectively characterize cellular density, perfusion, and heterogeneity, the HCCs were divided into four habitats. The volume fraction of each habitat was quantified. The logistic regression was used to explore the risk factors from habitat fraction and clinical variables. Clinical, habitat, and nomogram models were constructed using the identified risk factors from clinical characteristics, habitat fraction, and their combination, respectively. The diagnostic accuracy was evaluated using the area under the receiver operating characteristic curves (AUCs). Results MVI-positive HCC exhibited a significantly higher fraction of habitat 4 (f4) and a significantly lower fraction of habitat 2 (f2) (p < 0.001), which were selected as risk factors. Additionally, tumor size and elevated alpha-fetoprotein (AFP) were also included as risk factors for MVI. The nomogram model demonstrated the highest diagnostic performance (AUC = 0.807), followed by the habitat model (AUC = 0.777) and the clinical model (AUC = 0.708). Decision curve analysis indicated that the nomogram model offered more net benefit in identifying MVI compared to the clinical model. Conclusions DWI-based habitat imaging shows clinical potential for noninvasively and preoperatively determining the MVI of HCC with high accuracy. Clinical relevance statement The proposed strategy, diffusion-weighted imaging–based habitat imaging, can be applied for preoperatively and noninvasively identifying microvascular invasion in hepatocellular carcinoma, which offers potential benefits in terms of prognostic prediction and clinical management. Key Points • This study proposed a strategy of DWI-based habitat imaging for hepatocellular carcinoma. • The habitat imaging–derived metrics can serve as diagnostic markers for identifying the microvascular invasion. • Integrating the habitat-based metric and clinical variable, a predictive nomogram was constructed and displayed high accuracy for predicting microvascular invasion.