Jun Cheng
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
Jun Cheng has a total of 121 co-authors. Here are the top co-authors by number of shared publications:
- Dong Ni
Shenzhen University
12 shared publications Wufeng Xue
National‐Regional Key Technology Engineering Laboratory for Medical Ultrasound, Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging, School of Biomedical Engineering, Thoracic Surgery Department of the First Affiliated Hospital Shenzhen University Medical School, Shenzhen University Shenzhen PR China
6 shared publicationsDong Ni
Shenzhen University Medical School, Shenzhen University
4 shared publicationsWenhui Hong
National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging, School of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, China
4 shared publicationsHan Huang
Shenzhen University Medical School, Shenzhen University
3 shared publications- Jie Zhang
Department of Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, Indiana, USA
3 shared publications
Latest Publications
- Open Access
- 18 September 2025
- Published by Wiley in The Journal of Pathology: Clinical Research
- Vol. 11 (5) , e70047
Automated multi‐regional IHC scoring enhances prognostication in colorectal cancer
The tumor immune microenvironment plays a critical role in colorectal cancer (CRC) prognosis. However, most studies assess a limited set of immune markers manually in the tumor region, without considering immune heterogeneity across multiple tissue regions. This study aims to enhance CRC prognostic assessment by developing an automated multi‐regional immunohistochemistry (IHC) scoring system for 15 immune markers. Two representative tissue cores were extracted from CRC surgical specimens (n = 154) across four regions: tumor center, invasive margin, paracancerous tissues, and normal tissues. IHC staining was performed for 15 immune markers, and digitized slides were analyzed using computational algorithms to classify tissue types (e.g., glands, tumor, and stroma) and identify stained pixels. Immune infiltration was quantified in different tissue types across regions, and a tumor‐to‐healthy immune ratio (THIR) score was introduced to compare immune marker expression in tumor versus healthy stroma. Associations between IHC scores and overall survival (OS) and relapse‐free survival (RFS) were evaluated. Computational models achieved 95.19% accuracy in tissue classification and 97.90% in staining identification. Analysis of 120 IHC scores (15 markers × 8 tissue types) revealed significant immune heterogeneity, with 56 scores correlating with OS and 54 with RFS. Notably, markers such as Granzyme B and CD4 had higher prognostic relevance at the invasive margin than the tumor center, while markers like S100 and CD20 exhibited opposing prognostic effects across regions. Integrating multiple markers significantly improved prognostic accuracy, with the combined marker score in normal stroma providing the most significant risk stratification (log‐rank test, p = 1.56e‐7, OS). The THIR score also strongly correlated with patient outcomes. This study advances CRC prognostication through automated multi‐regional IHC scoring, highlighting the importance of immune heterogeneity across tissue regions. These findings support integrating region‐specific immune profiling into clinical workflows for more personalized and precise patient care.
- 18 September 2025
- Published by Wiley in The Journal of Pathology: Clinical Research
- Vol. 11 (5) , e70047
Open AccessAutomated multi‐regional IHC scoring enhances prognostication in colorectal cancer
The tumor immune microenvironment plays a critical role in colorectal cancer (CRC) prognosis. However, most studies assess a limited set of immune markers manually in the tumor region, without considering immune heterogeneity across multiple tissue regions. This study aims to enhance CRC prognostic assessment by developing an automated multi‐regional immunohistochemistry (IHC) scoring system for 15 immune markers. Two representative tissue cores were extracted from CRC surgical specimens (n = 154) across four regions: tumor center, invasive margin, paracancerous tissues, and normal tissues. IHC staining was performed for 15 immune markers, and digitized slides were analyzed using computational algorithms to classify tissue types (e.g., glands, tumor, and stroma) and identify stained pixels. Immune infiltration was quantified in different tissue types across regions, and a tumor‐to‐healthy immune ratio (THIR) score was introduced to compare immune marker expression in tumor versus healthy stroma. Associations between IHC scores and overall survival (OS) and relapse‐free survival (RFS) were evaluated. Computational models achieved 95.19% accuracy in tissue classification and 97.90% in staining identification. Analysis of 120 IHC scores (15 markers × 8 tissue types) revealed significant immune heterogeneity, with 56 scores correlating with OS and 54 with RFS. Notably, markers such as Granzyme B and CD4 had higher prognostic relevance at the invasive margin than the tumor center, while markers like S100 and CD20 exhibited opposing prognostic effects across regions. Integrating multiple markers significantly improved prognostic accuracy, with the combined marker score in normal stroma providing the most significant risk stratification (log‐rank test, p = 1.56e‐7, OS). The THIR score also strongly correlated with patient outcomes. This study advances CRC prognostication through automated multi‐regional IHC scoring, highlighting the importance of immune heterogeneity across tissue regions. These findings support integrating region‐specific immune profiling into clinical workflows for more personalized and precise patient care.
- 1 July 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (4) , e240625
Artificial Intelligence in Breast US Diagnosis and Report Generation
Using multimodality and multiview breast US images, the developed artificial intelligence system generated high-quality reports and helped junior radiologists improve the diagnosis of Breast Imaging Reporting and Data System categories.
- 1 July 2025
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 7 (4) , e240625
Artificial Intelligence in Breast US Diagnosis and Report Generation
Using multimodality and multiview breast US images, the developed artificial intelligence system generated high-quality reports and helped junior radiologists improve the diagnosis of Breast Imaging Reporting and Data System categories.
- Open Access
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 8, 0937
Noninvasive Computed Tomography-Based Quantification of Tumor Fibrosis Predicts Pancreatic Cancer Response to Gemcitabine/Nab-Paclitaxel
Pancreatic ductal adenocarcinoma (PDAC) carries a dismal prognosis. Chemotherapy remains the mainstay for unresectable cases, yet regimens like AG (gemcitabine/nab-paclitaxel) exhibit heterogeneous efficacy. Tumor fibrosis has emerged as a potential predictor of treatment response but lacks validated noninvasive assessment methods. To address this, in this multicenter study, tumor fibrosis was quantified in 361 patients with resectable PDAC from SYSUCC, XYCSU, and TCGA cohorts using deep learning-based tissue segmentation on hematoxylin and eosin-stained whole-slide images. Fibrosis was defined as stromal proportion, and its association with overall survival (OS) was evaluated. Transcriptomic profiling was performed in 51 XYCSU cases to validate the biological relevance of fibrosis quantification. A radiomics model (RM) was then developed using preoperative contrast-enhanced computed tomography (CT) scans from SYSUCC to predict fibrosis and externally validated in XYCSU. Clinical utility was assessed in an independent cohort of 295 unresectable PDAC patients treated with AG, FOLFIRINOX, or SOXIRI. High fibrosis correlated with prolonged OS across resectable cohorts (all P < 0.05). Transcriptomic analysis revealed enrichment of fibrosis-related pathways in high-fibrosis tumors. The RM achieved an area under the curve of 0.718 (95% confidence interval: 0.627 to 0.823) in the external test set. Among patients receiving AG, those with CT-predicted high fibrosis had significantly longer progression-free survival (median: 6.23 versus 4.70 months, P = 0.037) and OS (13.37 versus 7.73 months, P = 0.002). No significant survival benefit was observed for high-fibrosis patients receiving FOLFIRINOX or SOXIRI. CT-based fibrosis quantification offers a robust, noninvasive biomarker for predicting AG efficacy in unresectable PDAC.
- 1 January 2025
- Published by American Association for the Advancement of Science (AAAS) in Research
- Vol. 8, 0937
Open AccessNoninvasive Computed Tomography-Based Quantification of Tumor Fibrosis Predicts Pancreatic Cancer Response to Gemcitabine/Nab-Paclitaxel
Pancreatic ductal adenocarcinoma (PDAC) carries a dismal prognosis. Chemotherapy remains the mainstay for unresectable cases, yet regimens like AG (gemcitabine/nab-paclitaxel) exhibit heterogeneous efficacy. Tumor fibrosis has emerged as a potential predictor of treatment response but lacks validated noninvasive assessment methods. To address this, in this multicenter study, tumor fibrosis was quantified in 361 patients with resectable PDAC from SYSUCC, XYCSU, and TCGA cohorts using deep learning-based tissue segmentation on hematoxylin and eosin-stained whole-slide images. Fibrosis was defined as stromal proportion, and its association with overall survival (OS) was evaluated. Transcriptomic profiling was performed in 51 XYCSU cases to validate the biological relevance of fibrosis quantification. A radiomics model (RM) was then developed using preoperative contrast-enhanced computed tomography (CT) scans from SYSUCC to predict fibrosis and externally validated in XYCSU. Clinical utility was assessed in an independent cohort of 295 unresectable PDAC patients treated with AG, FOLFIRINOX, or SOXIRI. High fibrosis correlated with prolonged OS across resectable cohorts (all P < 0.05). Transcriptomic analysis revealed enrichment of fibrosis-related pathways in high-fibrosis tumors. The RM achieved an area under the curve of 0.718 (95% confidence interval: 0.627 to 0.823) in the external test set. Among patients receiving AG, those with CT-predicted high fibrosis had significantly longer progression-free survival (median: 6.23 versus 4.70 months, P = 0.037) and OS (13.37 versus 7.73 months, P = 0.002). No significant survival benefit was observed for high-fibrosis patients receiving FOLFIRINOX or SOXIRI. CT-based fibrosis quantification offers a robust, noninvasive biomarker for predicting AG efficacy in unresectable PDAC.
- 1 October 2023
- Published by Springer Nature
- p. 642-651
MUVF-YOLOX: A Multi-modal Ultrasound Video Fusion Network for Renal Tumor Diagnosis
- 1 October 2023
- Published by Springer Nature
- p. 642-651
MUVF-YOLOX: A Multi-modal Ultrasound Video Fusion Network for Renal Tumor Diagnosis
- Open Access
MUVF-YOLOX: A Multi-modal Ultrasound Video Fusion Network for Renal Tumor Diagnosis
Early diagnosis of renal cancer can greatly improve the survival rate of patients. Contrast-enhanced ultrasound (CEUS) is a cost-effective and non-invasive imaging technique and has become more and more frequently used for renal tumor diagnosis. However, the classification of benign and malignant renal tumors can still be very challenging due to the highly heterogeneous appearance of cancer and imaging artifacts. Our aim is to detect and classify renal tumors by integrating B-mode and CEUS-mode ultrasound videos. To this end, we propose a novel multi-modal ultrasound video fusion network that can effectively perform multi-modal feature fusion and video classification for renal tumor diagnosis. The attention-based multi-modal fusion module uses cross-attention and self-attention to extract modality-invariant features and modality-specific features in parallel. In addition, we design an object-level temporal aggregation (OTA) module that can automatically filter low-quality features and efficiently integrate temporal information from multiple frames to improve the accuracy of tumor diagnosis. Experimental results on a multicenter dataset show that the proposed framework outperforms the single-modal models and the competing methods. Furthermore, our OTA module achieves higher classification accuracy than the frame-level predictions. Our code is available at \url{https://github.com/JeunyuLi/MUAF}.
Open AccessMUVF-YOLOX: A Multi-modal Ultrasound Video Fusion Network for Renal Tumor Diagnosis
Early diagnosis of renal cancer can greatly improve the survival rate of patients. Contrast-enhanced ultrasound (CEUS) is a cost-effective and non-invasive imaging technique and has become more and more frequently used for renal tumor diagnosis. However, the classification of benign and malignant renal tumors can still be very challenging due to the highly heterogeneous appearance of cancer and imaging artifacts. Our aim is to detect and classify renal tumors by integrating B-mode and CEUS-mode ultrasound videos. To this end, we propose a novel multi-modal ultrasound video fusion network that can effectively perform multi-modal feature fusion and video classification for renal tumor diagnosis. The attention-based multi-modal fusion module uses cross-attention and self-attention to extract modality-invariant features and modality-specific features in parallel. In addition, we design an object-level temporal aggregation (OTA) module that can automatically filter low-quality features and efficiently integrate temporal information from multiple frames to improve the accuracy of tumor diagnosis. Experimental results on a multicenter dataset show that the proposed framework outperforms the single-modal models and the competing methods. Furthermore, our OTA module achieves higher classification accuracy than the frame-level predictions. Our code is available at \url{https://github.com/JeunyuLi/MUAF}.
- 16 September 2022
- Published by Springer Nature
- p. 665-674
Personalized Diagnostic Tool for Thyroid Cancer Classification Using Multi-view Ultrasound
- 16 September 2022
- Published by Springer Nature
- p. 665-674
Personalized Diagnostic Tool for Thyroid Cancer Classification Using Multi-view Ultrasound
- Open Access
- 6 September 2022
- Published by Frontiers Media SA in Frontiers in Molecular Biosciences
- Vol. 9, 982703
Multimodal ultrasound fusion network for differentiating between benign and malignant solid renal tumors
Objective: We aim to establish a deep learning model called multimodal ultrasound fusion network (MUF-Net) based on gray-scale and contrast-enhanced ultrasound (CEUS) images for classifying benign and malignant solid renal tumors automatically and to compare the model’s performance with the assessments by radiologists with different levels of experience.Methods: A retrospective study included the CEUS videos of 181 patients with solid renal tumors (81 benign and 100 malignant tumors) from June 2012 to June 2021. A total of 9794 B-mode and CEUS-mode images were cropped from the CEUS videos. The MUF-Net was proposed to combine gray-scale and CEUS images to differentiate benign and malignant solid renal tumors. In this network, two independent branches were designed to extract features from each of the two modalities, and the features were fused using adaptive weights. Finally, the network output a classification score based on the fused features. The model’s performance was evaluated using five-fold cross-validation and compared with the assessments of the two groups of radiologists with different levels of experience.Results: For the discrimination between benign and malignant solid renal tumors, the junior radiologist group, senior radiologist group, and MUF-Net achieved accuracy of 70.6%, 75.7%, and 80.0%, sensitivity of 89.3%, 95.9%, and 80.4%, specificity of 58.7%, 62.9%, and 79.1%, and area under the receiver operating characteristic curve of 0.740 (95% confidence internal (CI): 0.70–0.75), 0.794 (95% CI: 0.72–0.83), and 0.877 (95% CI: 0.83–0.93), respectively.Conclusion: The MUF-Net model can accurately classify benign and malignant solid renal tumors and achieve better performance than senior radiologists.Key points: The CEUS video data contain the entire tumor microcirculation perfusion characteristics. The proposed MUF-Net based on B-mode and CEUS-mode images can accurately distinguish between benign and malignant solid renal tumors with an area under the receiver operating characteristic curve of 0.877, which surpasses senior radiologists’ assessments by a large margin.
- 6 September 2022
- Published by Frontiers Media SA in Frontiers in Molecular Biosciences
- Vol. 9, 982703
Open AccessMultimodal ultrasound fusion network for differentiating between benign and malignant solid renal tumors
Objective: We aim to establish a deep learning model called multimodal ultrasound fusion network (MUF-Net) based on gray-scale and contrast-enhanced ultrasound (CEUS) images for classifying benign and malignant solid renal tumors automatically and to compare the model’s performance with the assessments by radiologists with different levels of experience.Methods: A retrospective study included the CEUS videos of 181 patients with solid renal tumors (81 benign and 100 malignant tumors) from June 2012 to June 2021. A total of 9794 B-mode and CEUS-mode images were cropped from the CEUS videos. The MUF-Net was proposed to combine gray-scale and CEUS images to differentiate benign and malignant solid renal tumors. In this network, two independent branches were designed to extract features from each of the two modalities, and the features were fused using adaptive weights. Finally, the network output a classification score based on the fused features. The model’s performance was evaluated using five-fold cross-validation and compared with the assessments of the two groups of radiologists with different levels of experience.Results: For the discrimination between benign and malignant solid renal tumors, the junior radiologist group, senior radiologist group, and MUF-Net achieved accuracy of 70.6%, 75.7%, and 80.0%, sensitivity of 89.3%, 95.9%, and 80.4%, specificity of 58.7%, 62.9%, and 79.1%, and area under the receiver operating characteristic curve of 0.740 (95% confidence internal (CI): 0.70–0.75), 0.794 (95% CI: 0.72–0.83), and 0.877 (95% CI: 0.83–0.93), respectively.Conclusion: The MUF-Net model can accurately classify benign and malignant solid renal tumors and achieve better performance than senior radiologists.Key points: The CEUS video data contain the entire tumor microcirculation perfusion characteristics. The proposed MUF-Net based on B-mode and CEUS-mode images can accurately distinguish between benign and malignant solid renal tumors with an area under the receiver operating characteristic curve of 0.877, which surpasses senior radiologists’ assessments by a large margin.
- Open Access
- 25 July 2022
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 13
Deep learning predicts immune checkpoint inhibitor-related pneumonitis from pretreatment computed tomography images
Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of lung cancer, including both non-small cell lung cancer and small cell lung cancer. Despite the promising results of immunotherapies, ICI-related pneumonitis (ICIP) is a potentially fatal adverse event. Therefore, early detection of patients at risk for developing ICIP before the initiation of immunotherapy is critical for alleviating future complications with early interventions and improving treatment outcomes. In this study, we present the first reported work that explores the potential of deep learning to predict patients who are at risk for developing ICIP. To this end, we collected the pretreatment baseline CT images and clinical information of 24 patients who developed ICIP after immunotherapy and 24 control patients who did not. A multimodal deep learning model was constructed based on 3D CT images and clinical data. To enhance performance, we employed two-stage transfer learning by pre-training the model sequentially on a large natural image dataset and a large CT image dataset, as well as transfer learning. Extensive experiments were conducted to verify the effectiveness of the key components used in our method. Using five-fold cross-validation, our method accurately distinguished ICIP patients from non-ICIP patients, with area under the receiver operating characteristic curve of 0.918 and accuracy of 0.920. This study demonstrates the promising potential of deep learning to identify patients at risk for developing ICIP. The proposed deep learning model enables efficient risk stratification, close monitoring, and prompt management of ICIP, ultimately leading to better treatment outcomes.
- 25 July 2022
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 13
Open AccessDeep learning predicts immune checkpoint inhibitor-related pneumonitis from pretreatment computed tomography images
Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of lung cancer, including both non-small cell lung cancer and small cell lung cancer. Despite the promising results of immunotherapies, ICI-related pneumonitis (ICIP) is a potentially fatal adverse event. Therefore, early detection of patients at risk for developing ICIP before the initiation of immunotherapy is critical for alleviating future complications with early interventions and improving treatment outcomes. In this study, we present the first reported work that explores the potential of deep learning to predict patients who are at risk for developing ICIP. To this end, we collected the pretreatment baseline CT images and clinical information of 24 patients who developed ICIP after immunotherapy and 24 control patients who did not. A multimodal deep learning model was constructed based on 3D CT images and clinical data. To enhance performance, we employed two-stage transfer learning by pre-training the model sequentially on a large natural image dataset and a large CT image dataset, as well as transfer learning. Extensive experiments were conducted to verify the effectiveness of the key components used in our method. Using five-fold cross-validation, our method accurately distinguished ICIP patients from non-ICIP patients, with area under the receiver operating characteristic curve of 0.918 and accuracy of 0.920. This study demonstrates the promising potential of deep learning to identify patients at risk for developing ICIP. The proposed deep learning model enables efficient risk stratification, close monitoring, and prompt management of ICIP, ultimately leading to better treatment outcomes.
- Open Access
- 15 June 2022
- Published by Springer Nature in Journal of Translational Medicine
- Vol. 20 (1) , 1-13
Multimodal data analysis reveals that pancreatobiliary-type ampullary adenocarcinoma resembles pancreatic adenocarcinoma and differs from cholangiocarcinoma
Background: Ampullary adenocarcinoma (AAC) arises from the ampulla of Vater where the pancreatic duct and bile duct join and empty into the duodenum. It can be classified into intestinal and pancreatobiliary types based on histopathology or immunohistochemistry. However, there are no biomarkers for further classification of pancreatobiliary-type AAC which has important implications for its treatment. We aimed to identify the tumor origin of pancreatobiliary-type AAC by systematically analyzing whole-slide images (WSIs), survival data, and genome sequencing data collected from multiple centers.Methods: This study involved three experiments. First, we extracted quantitative and highly interpretable features from the tumor region in WSIs and constructed a histologic classifier to differentiate between pancreatic adenocarcinoma (PAC) and cholangiocarcinoma. The histologic classifier was then applied to patients with pancreatobiliary-type AAC to infer the tumor origin. Secondly, we compared the overall survival of patients with pancreatobiliary-type AAC stratified by the adjuvant chemotherapy regimens designed for PAC or cholangiocarcinoma. Finally, we compared the mutation landscape of pancreatobiliary-type AAC with those of PAC and cholangiocarcinoma.Results: The histologic classifier accurately classified PAC and cholangiocarcinoma in both the internal and external validation sets (AUC > 0.99). All pancreatobiliary-type AACs (n = 45) were classified as PAC. The patients with pancreatobiliary-type AAC receiving regimens designed for PAC showed more favorable overall survival than those receiving regimens designed for cholangiocarcinoma in a multivariable Cox regression (hazard ratio = 7.24, 95% confidence interval: 1.28–40.78,P = 0.025). The results of mutation analysis showed that the mutation landscape of AAC was very similar to that of PAC but distinct from that of cholangiocarcinoma.Conclusions: This multi-center study provides compelling evidence that pancreatobiliary-type AAC resembles PAC instead of cholangiocarcinoma in different aspects, which can guide the treatment selection and clinical trials planning for pancreatobiliary-type AAC.
- 15 June 2022
- Published by Springer Nature in Journal of Translational Medicine
- Vol. 20 (1) , 1-13
Open AccessMultimodal data analysis reveals that pancreatobiliary-type ampullary adenocarcinoma resembles pancreatic adenocarcinoma and differs from cholangiocarcinoma
Background: Ampullary adenocarcinoma (AAC) arises from the ampulla of Vater where the pancreatic duct and bile duct join and empty into the duodenum. It can be classified into intestinal and pancreatobiliary types based on histopathology or immunohistochemistry. However, there are no biomarkers for further classification of pancreatobiliary-type AAC which has important implications for its treatment. We aimed to identify the tumor origin of pancreatobiliary-type AAC by systematically analyzing whole-slide images (WSIs), survival data, and genome sequencing data collected from multiple centers.Methods: This study involved three experiments. First, we extracted quantitative and highly interpretable features from the tumor region in WSIs and constructed a histologic classifier to differentiate between pancreatic adenocarcinoma (PAC) and cholangiocarcinoma. The histologic classifier was then applied to patients with pancreatobiliary-type AAC to infer the tumor origin. Secondly, we compared the overall survival of patients with pancreatobiliary-type AAC stratified by the adjuvant chemotherapy regimens designed for PAC or cholangiocarcinoma. Finally, we compared the mutation landscape of pancreatobiliary-type AAC with those of PAC and cholangiocarcinoma.Results: The histologic classifier accurately classified PAC and cholangiocarcinoma in both the internal and external validation sets (AUC > 0.99). All pancreatobiliary-type AACs (n = 45) were classified as PAC. The patients with pancreatobiliary-type AAC receiving regimens designed for PAC showed more favorable overall survival than those receiving regimens designed for cholangiocarcinoma in a multivariable Cox regression (hazard ratio = 7.24, 95% confidence interval: 1.28–40.78,P = 0.025). The results of mutation analysis showed that the mutation landscape of AAC was very similar to that of PAC but distinct from that of cholangiocarcinoma.Conclusions: This multi-center study provides compelling evidence that pancreatobiliary-type AAC resembles PAC instead of cholangiocarcinoma in different aspects, which can guide the treatment selection and clinical trials planning for pancreatobiliary-type AAC.