Fubi Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Fubi Hu has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Zixing Huang
West China Hospital 3 Department of Radiology , Chengdu, China
4 shared publications - Bin Song
Department of Radiology, West China Hospital, Sichuan University, Chengdu, China
3 shared publications - Hongkun Yin
Department of Radiology, West China Hospital, Sichuan University, Chengdu, China
3 shared publications Ru Yang
Department of Radiology, The First Affiliated Hospital of Chengdu Medical College, No.278, Baoguang Road, Xindu District, Chengdu, Sichuan, China (R.Y., J.P., F.H.).
3 shared publications- Dan Liu
Department of Radiology, West China Hospital, Sichuan University, Chengdu, China
2 shared publications - Huiling Zhang
Institute of Advanced Research, Infervision Medical Technology Co Ltd, Beijing, China
2 shared publications
Latest Publications
- 8 August 2025
- Published by Wiley in Liver International
- Vol. 45 (9) , e70274
Deep Learning‐Enhanced CTA for Noninvasive Prediction of First Variceal Haemorrhage in Cirrhosis: A Multi‐Centre Study
Background and Aims: The first variceal haemorrhage (FVH) is a life‐threatening complication of liver cirrhosis that requires timely intervention; however, noninvasive tools for accurately predicting FVH remain limited. This study aimed to develop noninvasive, deep learning‐enhanced computed tomographic angiography (CTA) models for early and accurate FVH prediction. Methods: This multi‐centre retrospective study included 184 cirrhotic patients (FVH: n = 107, non‐FVH: n = 77) enrolled from December 2014 to May 2022. Patients were randomly divided (7:3) into training and validation cohorts. CTA and clinical data were collected and analysed. A novel Vision‐Transformer (ViT) network, combined with reinforcement learning (RL), was applied to CTA images to predict FVH and was compared with convolutional neural networks (CNNs). Models were evaluated using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA), and feature importance was determined from model coefficients and gradients. Results: The ViT + RL* model demonstrated superior diagnostic performance, achieving an AUC of 0.985 (95% CI, 0.955–1.0) in the validation cohort and 0.956 (95% CI, 0.919–0.988) in the training cohort, outperforming traditional CNNs. DCA and the area under the curve confirmed the enhanced clinical utility of the ViT + RL* model compared to CNNs; the ViT + RL* model highlighted critical regions in the liver, spleen, oesophageal lumen, and abdominal vessels. Meanwhile, clinical data identified creatinine and prothrombin time as potential predictive factors, with moderate predictive performance. Conclusions: The novel deep learning‐enhanced CTA models offer a robust, non‐invasive method for predicting FVH, with the ViT + RL* model demonstrating excellent efficacy; thus providing a valuable tool for early risk stratification in cirrhotic patients.
- 8 August 2025
- Published by Wiley in Liver International
- Vol. 45 (9) , e70274
Deep Learning‐Enhanced CTA for Noninvasive Prediction of First Variceal Haemorrhage in Cirrhosis: A Multi‐Centre Study
Background and Aims: The first variceal haemorrhage (FVH) is a life‐threatening complication of liver cirrhosis that requires timely intervention; however, noninvasive tools for accurately predicting FVH remain limited. This study aimed to develop noninvasive, deep learning‐enhanced computed tomographic angiography (CTA) models for early and accurate FVH prediction. Methods: This multi‐centre retrospective study included 184 cirrhotic patients (FVH: n = 107, non‐FVH: n = 77) enrolled from December 2014 to May 2022. Patients were randomly divided (7:3) into training and validation cohorts. CTA and clinical data were collected and analysed. A novel Vision‐Transformer (ViT) network, combined with reinforcement learning (RL), was applied to CTA images to predict FVH and was compared with convolutional neural networks (CNNs). Models were evaluated using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA), and feature importance was determined from model coefficients and gradients. Results: The ViT + RL* model demonstrated superior diagnostic performance, achieving an AUC of 0.985 (95% CI, 0.955–1.0) in the validation cohort and 0.956 (95% CI, 0.919–0.988) in the training cohort, outperforming traditional CNNs. DCA and the area under the curve confirmed the enhanced clinical utility of the ViT + RL* model compared to CNNs; the ViT + RL* model highlighted critical regions in the liver, spleen, oesophageal lumen, and abdominal vessels. Meanwhile, clinical data identified creatinine and prothrombin time as potential predictive factors, with moderate predictive performance. Conclusions: The novel deep learning‐enhanced CTA models offer a robust, non‐invasive method for predicting FVH, with the ViT + RL* model demonstrating excellent efficacy; thus providing a valuable tool for early risk stratification in cirrhotic patients.
- 11 April 2025
- Published by Elsevier in Academic Radiology
- Vol. 32 (8) , 4524-4531
A High-resolution T2WI-based Deep Learning Model for Preoperative Discrimination Between T2 and T3 Rectal Cancer: A Multicenter Study
- 11 April 2025
- Published by Elsevier in Academic Radiology
- Vol. 32 (8) , 4524-4531
A High-resolution T2WI-based Deep Learning Model for Preoperative Discrimination Between T2 and T3 Rectal Cancer: A Multicenter Study
- Open Access
- 9 October 2024
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e56851
Development and Validation of a Computed Tomography–Based Model for Noninvasive Prediction of the T Stage in Gastric Cancer: Multicenter Retrospective Study
Background: As part of the TNM (tumor-node-metastasis) staging system, T staging based on tumor depth is crucial for developing treatment plans. Previous studies have constructed a deep learning model based on computed tomographic (CT) radiomic signatures to predict the number of lymph node metastases and survival in patients with resected gastric cancer (GC). However, few studies have reported the combination of deep learning and radiomics in predicting T staging in GC. Objective: This study aimed to develop a CT-based model for automatic prediction of the T stage of GC via radiomics and deep learning. Methods: A total of 771 GC patients from 3 centers were retrospectively enrolled and divided into training, validation, and testing cohorts. Patients with GC were classified into mild (stage T1 and T2), moderate (stage T3), and severe (stage T4) groups. Three predictive models based on the labeled CT images were constructed using the radiomics features (radiomics model), deep features (deep learning model), and a combination of both (hybrid model). Results: The overall classification accuracy of the radiomics model was 64.3% in the internal testing data set. The deep learning model and hybrid model showed better performance than the radiomics model, with overall classification accuracies of 75.7% (P=.04) and 81.4% (P=.001), respectively. On the subtasks of binary classification of tumor severity, the areas under the curve of the radiomics, deep learning, and hybrid models were 0.875, 0.866, and 0.886 in the internal testing data set and 0.820, 0.818, and 0.972 in the external testing data set, respectively, for differentiating mild (stage T1~T2) from nonmild (stage T3~T4) patients, and were 0.815, 0.892, and 0.894 in the internal testing data set and 0.685, 0.808, and 0.897 in the external testing data set, respectively, for differentiating nonsevere (stage T1~T3) from severe (stage T4) patients. Conclusions: The hybrid model integrating radiomics features and deep features showed favorable performance in diagnosing the pathological stage of GC.
- 9 October 2024
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 26, e56851
Open AccessDevelopment and Validation of a Computed Tomography–Based Model for Noninvasive Prediction of the T Stage in Gastric Cancer: Multicenter Retrospective Study
Background: As part of the TNM (tumor-node-metastasis) staging system, T staging based on tumor depth is crucial for developing treatment plans. Previous studies have constructed a deep learning model based on computed tomographic (CT) radiomic signatures to predict the number of lymph node metastases and survival in patients with resected gastric cancer (GC). However, few studies have reported the combination of deep learning and radiomics in predicting T staging in GC. Objective: This study aimed to develop a CT-based model for automatic prediction of the T stage of GC via radiomics and deep learning. Methods: A total of 771 GC patients from 3 centers were retrospectively enrolled and divided into training, validation, and testing cohorts. Patients with GC were classified into mild (stage T1 and T2), moderate (stage T3), and severe (stage T4) groups. Three predictive models based on the labeled CT images were constructed using the radiomics features (radiomics model), deep features (deep learning model), and a combination of both (hybrid model). Results: The overall classification accuracy of the radiomics model was 64.3% in the internal testing data set. The deep learning model and hybrid model showed better performance than the radiomics model, with overall classification accuracies of 75.7% (P=.04) and 81.4% (P=.001), respectively. On the subtasks of binary classification of tumor severity, the areas under the curve of the radiomics, deep learning, and hybrid models were 0.875, 0.866, and 0.886 in the internal testing data set and 0.820, 0.818, and 0.972 in the external testing data set, respectively, for differentiating mild (stage T1~T2) from nonmild (stage T3~T4) patients, and were 0.815, 0.892, and 0.894 in the internal testing data set and 0.685, 0.808, and 0.897 in the external testing data set, respectively, for differentiating nonsevere (stage T1~T3) from severe (stage T4) patients. Conclusions: The hybrid model integrating radiomics features and deep features showed favorable performance in diagnosing the pathological stage of GC.
- Open Access
- 31 January 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Development and Validation of a Computed Tomography–Based Model for Noninvasive Prediction of the T Stage in Gastric Cancer: Multicenter Retrospective Study (Preprint)
BACKGROUND As part of the TNM (tumor-node-metastasis) staging system, T staging based on tumor depth is crucial for developing treatment plans. Previous studies have constructed a deep learning model based on computed tomographic (CT) radiomic signatures to predict the number of lymph node metastases and survival in patients with resected gastric cancer (GC). However, few studies have reported the combination of deep learning and radiomics in predicting T staging in GC. OBJECTIVE This study aimed to develop a CT-based model for automatic prediction of the T stage of GC via radiomics and deep learning. METHODS A total of 771 GC patients from 3 centers were retrospectively enrolled and divided into training, validation, and testing cohorts. Patients with GC were classified into mild (stage T1 and T2), moderate (stage T3), and severe (stage T4) groups. Three predictive models based on the labeled CT images were constructed using the radiomics features (radiomics model), deep features (deep learning model), and a combination of both (hybrid model). RESULTS The overall classification accuracy of the radiomics model was 64.3% in the internal testing data set. The deep learning model and hybrid model showed better performance than the radiomics model, with overall classification accuracies of 75.7% (P=.04) and 81.4% (P=.001), respectively. On the subtasks of binary classification of tumor severity, the areas under the curve of the radiomics, deep learning, and hybrid models were 0.875, 0.866, and 0.886 in the internal testing data set and 0.820, 0.818, and 0.972 in the external testing data set, respectively, for differentiating mild (stage T1~T2) from nonmild (stage T3~T4) patients, and were 0.815, 0.892, and 0.894 in the internal testing data set and 0.685, 0.808, and 0.897 in the external testing data set, respectively, for differentiating nonsevere (stage T1~T3) from severe (stage T4) patients. CONCLUSIONS The hybrid model integrating radiomics features and deep features showed favorable performance in diagnosing the pathological stage of GC.
- 31 January 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDevelopment and Validation of a Computed Tomography–Based Model for Noninvasive Prediction of the T Stage in Gastric Cancer: Multicenter Retrospective Study (Preprint)
BACKGROUND As part of the TNM (tumor-node-metastasis) staging system, T staging based on tumor depth is crucial for developing treatment plans. Previous studies have constructed a deep learning model based on computed tomographic (CT) radiomic signatures to predict the number of lymph node metastases and survival in patients with resected gastric cancer (GC). However, few studies have reported the combination of deep learning and radiomics in predicting T staging in GC. OBJECTIVE This study aimed to develop a CT-based model for automatic prediction of the T stage of GC via radiomics and deep learning. METHODS A total of 771 GC patients from 3 centers were retrospectively enrolled and divided into training, validation, and testing cohorts. Patients with GC were classified into mild (stage T1 and T2), moderate (stage T3), and severe (stage T4) groups. Three predictive models based on the labeled CT images were constructed using the radiomics features (radiomics model), deep features (deep learning model), and a combination of both (hybrid model). RESULTS The overall classification accuracy of the radiomics model was 64.3% in the internal testing data set. The deep learning model and hybrid model showed better performance than the radiomics model, with overall classification accuracies of 75.7% (P=.04) and 81.4% (P=.001), respectively. On the subtasks of binary classification of tumor severity, the areas under the curve of the radiomics, deep learning, and hybrid models were 0.875, 0.866, and 0.886 in the internal testing data set and 0.820, 0.818, and 0.972 in the external testing data set, respectively, for differentiating mild (stage T1~T2) from nonmild (stage T3~T4) patients, and were 0.815, 0.892, and 0.894 in the internal testing data set and 0.685, 0.808, and 0.897 in the external testing data set, respectively, for differentiating nonsevere (stage T1~T3) from severe (stage T4) patients. CONCLUSIONS The hybrid model integrating radiomics features and deep features showed favorable performance in diagnosing the pathological stage of GC.
- Open Access
- 3 December 2021
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 11
A Bounding Box-Based Radiomics Model for Detecting Occult Peritoneal Metastasis in Advanced Gastric Cancer: A Multicenter Study
Purpose: To develop a bounding box (BBOX)-based radiomics model for the preoperative diagnosis of occult peritoneal metastasis (OPM) in advanced gastric cancer (AGC) patients.Materials and Methods: 599 AGC patients from 3 centers were retrospectively enrolled and were divided into training, validation, and testing cohorts. The minimum circumscribed rectangle of the ROIs for the largest tumor area (R_BBOX), the nonoverlapping area between the tumor and R_BBOX (peritumoral area; PERI) and the smallest rectangle that could completely contain the tumor determined by a radiologist (M_BBOX) were used as inputs to extract radiomic features. Multivariate logistic regression was used to construct a radiomics model to estimate the preoperative probability of OPM in AGC patients.Results: The M_BBOX model was not significantly different from R_BBOX in the validation cohort [AUC: M_BBOX model 0.871 (95% CI, 0.814–0.940) vs. R_BBOX model 0.873 (95% CI, 0.820–0.940); p = 0.937]. M_BBOX was selected as the final radiomics model because of its extremely low annotation cost and superior OPM discrimination performance (sensitivity of 85.7% and specificity of 82.8%) over the clinical model, and this radiomics model showed comparable diagnostic efficacy in the testing cohort.Conclusions: The BBOX-based radiomics could serve as a simpler reliable and powerful tool for the preoperative diagnosis of OPM in AGC patients. And M_BBOX-based radiomics is simpler and less time consuming.
- 3 December 2021
- Published by Frontiers Media SA in Frontiers in Oncology
- Vol. 11
Open AccessA Bounding Box-Based Radiomics Model for Detecting Occult Peritoneal Metastasis in Advanced Gastric Cancer: A Multicenter Study
Purpose: To develop a bounding box (BBOX)-based radiomics model for the preoperative diagnosis of occult peritoneal metastasis (OPM) in advanced gastric cancer (AGC) patients.Materials and Methods: 599 AGC patients from 3 centers were retrospectively enrolled and were divided into training, validation, and testing cohorts. The minimum circumscribed rectangle of the ROIs for the largest tumor area (R_BBOX), the nonoverlapping area between the tumor and R_BBOX (peritumoral area; PERI) and the smallest rectangle that could completely contain the tumor determined by a radiologist (M_BBOX) were used as inputs to extract radiomic features. Multivariate logistic regression was used to construct a radiomics model to estimate the preoperative probability of OPM in AGC patients.Results: The M_BBOX model was not significantly different from R_BBOX in the validation cohort [AUC: M_BBOX model 0.871 (95% CI, 0.814–0.940) vs. R_BBOX model 0.873 (95% CI, 0.820–0.940); p = 0.937]. M_BBOX was selected as the final radiomics model because of its extremely low annotation cost and superior OPM discrimination performance (sensitivity of 85.7% and specificity of 82.8%) over the clinical model, and this radiomics model showed comparable diagnostic efficacy in the testing cohort.Conclusions: The BBOX-based radiomics could serve as a simpler reliable and powerful tool for the preoperative diagnosis of OPM in AGC patients. And M_BBOX-based radiomics is simpler and less time consuming.
- 1 June 2021
- Published by Elsevier in International Journal of Cardiology
- Vol. 332, 8-14
Relationship of epicardial fat volume with coronary plaque characteristics, coronary artery calcification score, coronary stenosis, and CT-FFR for lesion-specific ischemia in patients with known or suspected coronary artery disease
- 1 June 2021
- Published by Elsevier in International Journal of Cardiology
- Vol. 332, 8-14
Relationship of epicardial fat volume with coronary plaque characteristics, coronary artery calcification score, coronary stenosis, and CT-FFR for lesion-specific ischemia in patients with known or suspected coronary artery disease
- 17 June 2017
- Published by Springer Nature in Abdominal Radiology
- Vol. 42 (12) , 2855-2863
Liver fibrosis: in vivo evaluation using intravoxel incoherent motion-derived histogram metrics with histopathologic findings at 3.0 T
Purpose To retrospectively determine the feasibility of intravoxel incoherent motion (IVIM) imaging based on histogram analysis for the staging of liver fibrosis (LF) using histopathologic findings as the reference standard. Methods 56 consecutive patients (14 men, 42 women; age range, 15–76, years) with chronic liver diseases (CLDs) were studied using IVIM-DWI with 9 b-values (0, 25, 50, 75, 100, 150, 200, 500, 800 s/mm2) at 3.0 T. Fibrosis stage was evaluated using the METAVIR scoring system. Histogram metrics including mean, standard deviation (Std), skewness, kurtosis, minimum (Min), maximum (Max), range, interquartile (Iq) range, and percentiles (10, 25, 50, 75, 90th) were extracted from apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) maps. All histogram metrics among different fibrosis groups were compared using one-way analysis of variance or nonparametric Kruskal–Wallis test. For significant parameters, receivers operating characteristic curve (ROC) analyses were further performed for the staging of LF. Results Based on their METAVIR stage, the 56 patients were reclassified into three groups as follows: F0–1 group (n = 25), F2–3 group (n = 21), and F4 group (n = 10). The mean, Iq range, percentiles (50, 75, and 90th) of D* maps between the groups were significant differences (all P < 0.05). Area under the ROC curve (AUC) of the mean, Iq range, 50, 75, and 90th percentile of D* maps for identifying significant LF (≥F2 stage) was 0.901, 0.859, 0.876, 0.943, and 0.886 (all P < 0.0001), respectively; for diagnosing severe fibrosis or cirrhosis (F4), AUC was 0.917, 0.922, 0.943, 0.985, and 0.939 (all P < 0.0001), respectively. The histogram metrics of ADC, D, and f maps demonstrated no significant difference among the groups (all P > 0.05). Conclusion Histogram analysis of D* map derived from IVIM can be used to stage liver fibrosis in patients with CLDs and provide more quantitative information beyond the mean value.
- 17 June 2017
- Published by Springer Nature in Abdominal Radiology
- Vol. 42 (12) , 2855-2863
Liver fibrosis: in vivo evaluation using intravoxel incoherent motion-derived histogram metrics with histopathologic findings at 3.0 T
Purpose To retrospectively determine the feasibility of intravoxel incoherent motion (IVIM) imaging based on histogram analysis for the staging of liver fibrosis (LF) using histopathologic findings as the reference standard. Methods 56 consecutive patients (14 men, 42 women; age range, 15–76, years) with chronic liver diseases (CLDs) were studied using IVIM-DWI with 9 b-values (0, 25, 50, 75, 100, 150, 200, 500, 800 s/mm2) at 3.0 T. Fibrosis stage was evaluated using the METAVIR scoring system. Histogram metrics including mean, standard deviation (Std), skewness, kurtosis, minimum (Min), maximum (Max), range, interquartile (Iq) range, and percentiles (10, 25, 50, 75, 90th) were extracted from apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo-diffusion coefficient (D*), and perfusion fraction (f) maps. All histogram metrics among different fibrosis groups were compared using one-way analysis of variance or nonparametric Kruskal–Wallis test. For significant parameters, receivers operating characteristic curve (ROC) analyses were further performed for the staging of LF. Results Based on their METAVIR stage, the 56 patients were reclassified into three groups as follows: F0–1 group (n = 25), F2–3 group (n = 21), and F4 group (n = 10). The mean, Iq range, percentiles (50, 75, and 90th) of D* maps between the groups were significant differences (all P < 0.05). Area under the ROC curve (AUC) of the mean, Iq range, 50, 75, and 90th percentile of D* maps for identifying significant LF (≥F2 stage) was 0.901, 0.859, 0.876, 0.943, and 0.886 (all P < 0.0001), respectively; for diagnosing severe fibrosis or cirrhosis (F4), AUC was 0.917, 0.922, 0.943, 0.985, and 0.939 (all P < 0.0001), respectively. The histogram metrics of ADC, D, and f maps demonstrated no significant difference among the groups (all P > 0.05). Conclusion Histogram analysis of D* map derived from IVIM can be used to stage liver fibrosis in patients with CLDs and provide more quantitative information beyond the mean value.
- Open Access
- 1 January 2013
- Published by Baishideng Publishing Group Inc. in World Journal of Radiology
- Vol. 5 (4) , 187-192
Thoracic epidural angiolipoma: A case report and review of the literature
Thoracic epidural angiolipoma: A case report and review of the literature
- 1 January 2013
- Published by Baishideng Publishing Group Inc. in World Journal of Radiology
- Vol. 5 (4) , 187-192
Open AccessThoracic epidural angiolipoma: A case report and review of the literature
Thoracic epidural angiolipoma: A case report and review of the literature