YCYidi Chen

Yidi Chen

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yidi Chen has a total of 140 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    MRI-endoscopy deep learning for predicting pathologic complete response and disease-free survival in locally advanced rectal cancer

    by Yuanling Yang, Tianyu Wu, Yixin Liu, Lixuan Huang, Yiyun Xu, Yan Wu, Yifan Xu, Xinting Peng, Shenglian Wen, Yidi Chen, et al.

  • Uncertainty-Aware Cross-Modal Retrieval for Medical Report Generation

    by Nan Zhou, Meng Liu, Linchao He, Mengting Luo, Yidi Chen, Yi Zhang, Ke Zou, Hu Chen

    Automatic medical report generation (MRG) has advanced significantly with retrieval-augmented strategies. However, existing methods face two persistent challenges: 1) a largely reliance on single-modal retrieval, which limits multimodal semantic capture and cross-modal alignment; and 2) a lack of reliable information control, leading to irrelevant noisy content and potential hallucinations. To address these limitations, we propose Uncertainty-aware Cross-modal Alignment and Refinement, named U-CAR, a unified framework that enhances both semantic integration and retrieval reliability. First, a cross-modal alignment module explicitly learns fine-grained correspondences between visual and textual representations, ensuring consistent semantics across modalities. This alignment guides the construction of dual-path retrieval-aware memory banks, with one in the visual domain and one in the textual domain, enabling retrieval to capture complementary cues from both modalities. Second, we design a cross-modal retrieval-augmented generation strategy that jointly attends to the retrieved visual and textual context, thereby enriching semantic coverage and reinforcing the integration of multi-modal evidence in the generated reports. In parallel, we introduce an uncertainty-aware refinement mechanism that quantifies generation confidence to adaptively determine the necessity of retrieval. Experiments on the IU X-Ray and MIMIC-CXR datasets demonstrate that U-CAR outperforms the current state-of-the-art methods, achieving a 9% improvement in CIDEr on IU X-Ray. and a 4% gain in BLEU-4 on MIMIC-CXR. These results underscore U-CAR's effectiveness in generating accurate, coherent, and clinically relevant medical reports. Codes are available in https://github.com/Zhounan1222/U-CAR/tree/main.


  • Predictive model development and validation of functional liver imaging score for prognosis of patients with hepatocellular carcinoma after surgical resection: a multicenter study

    by Feier Ding, Takashi Ota, Shuo Cai, Hui Ma, Masahiro Yanagawa, Atsushi Nakamoto, Noriyuki Tomiyama, Yidi Chen, Bin Song, Xinya Zhao

    Purpose This study aimed to develop and validate a prognostic model for hepatocellular carcinoma (HCC) patients undergoing liver resection, using the functional liver imaging score (FLIS) derived from hepatobiliary-specific contrast-enhanced magnetic resonance imaging (MRI). Material and methods A total of 694 pathologically confirmed HCC patients who underwent hepatobiliary-specific MRI with either gadoxetic acid or gadobenate dimeglumine and subsequent liver resection were included. FLIS was calculated by assigning 0–2 points to three hepatobiliary-phase MRI features: hepatic enhancement, biliary excretion, and portal vein signal intensity. Multivariable Cox regression identified AFP level, tumor size, and extent of resection as independent predictors of overall survival (OS). Results FLIS ≤ 2, alpha-fetoprotein (AFP) > 400 ng/mL, tumor size > 5 cm, and major resection were identified as independent predictors of worse OS. A predictive model combining these factors demonstrated excellent prognostic performance, with Harrell’s concordance indices of 0.91 in the training cohort and 0.96 internal validation cohort, and 0.94 in external validation cohort. The FLIS-based model significantly outperformed FLIS alone and conventional clinical models (p < 0.05). Kaplan–Meier survival analysis showed that low-risk patients had significantly better OS and recurrence-free survival (RFS) compared to high-risk patients across all cohorts (p < 0.05). Conclusion FLIS is a simple, non-invasive imaging biomarker for evaluating liver function and predicting outcomes in HCC patients. When integrated with key clinical variables, the FLIS-based model demonstrates excellent discrimination and calibration for OS and RFS, providing accurate postoperative prognostic stratification and showing great potential for guiding surveillance and improving long-term survival outcomes in future clinical applications.


  • Open Access

    GEMeX-RMCoT: An Enhanced Med-VQA Dataset for Region-Aware Multimodal Chain-of-Thought Reasoning

    by Bo Liu, Xiangyu Zhao, Along He, Yidi Chen, Huazhu Fu, Xiao-Ming Wu


  • Open Access

    Dual-layer spectral detector computed tomography in distinguishing between bland and neoplastic portal vein thrombosis

    by Lin Zhang, Tianying Zheng, Mao Su, Xiaodi Zhang, Haiwei Liu, Bin Song, Yidi Chen

    Background: Differentiating bland from neoplastic portal vein thrombosis (PVT) is crucial for staging and treatment decisions in patients with suspected or confirmed liver malignancies. This study aimed to assess the diagnostic efficacy of the quantitative parameters of dual-layer spectral detector computed tomography (DLCT) in distinguishing between bland and neoplastic PVT. Methods: This single-center prospective study included consecutive patients with identifiable PVT who underwent contrast-enhanced liver DLCT between April 2022 and August 2023. The reference standard was established based on imaging, patient history, and follow-up data. Quantitative parameters, including computed tomography (CT) attenuation values from conventional and virtual monoenergetic (40–90 keV) images, iodine density, effective atomic number (Zeff), and arterial enhancement fraction (AEF), were documented. Interobserver agreement was assessed via the intraclass correlation coefficient (ICC). A combined score integrating DLCT quantitative parameters was developed, and diagnostic performance was evaluated according to the area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, and specificity. Results: A total of 81 patients (median age 54 years; 64 males) were enrolled, including 31 patients with bland PVT and 50 with neoplastic PVT. Baseline characteristics showed significant differences between the bland PVT and neoplastic PVT groups, including in α-fetoprotein (AFP) level (median 3.3 vs. 932.0 ng/mL, Pvs. 68.5 U/L, P=0.004), and PVT diameter (mean 13.6 vs. 16.5 mm, P=0.02). The quantitative DLCT parameters were significantly different between the bland and neoplastic groups (all P values ≤0.001). Interobserver agreement was good to excellent (ICC 0.796–0.973). Quantitative parameters from the arterial phase, as compared to those from the portal venous phase, exhibited higher diagnostic AUC values and specificity, but lower sensitivity for neoplastic PVT. The CT attenuation values of arterial-phase 40-keV virtual monoenergetic images achieved the highest AUC of 0.938 [95% confidence interval (CI): 0.862–0.980], with a sensitivity of 84.0% and a specificity of 96.8%. The AEF also showed high diagnostic performance, with an AUC of 0.930 (95% CI: 0.851–0.975), a sensitivity of 90.0%, and a specificity of 93.6%. The combined score derived from the arterial-phase DLCT quantitative parameters and AEF achieved the highest AUC of 0.979 (95% CI: 0.919–0.998) for diagnosing neoplastic PVT, with a sensitivity and specificity of 94.0% and 100.0%, respectively. Conclusions: DLCT quantitative parameters effectively distinguished between bland and neoplastic PVT. The combined score derived from quantitative DLCT parameters demonstrated superior diagnostic performance, potentially aiding in accurate staging and treatment decisions for patients with suspected or confirmed liver malignancies. Further large-scale multicenter studies are needed to confirm these findings.


  • Open Access

    From texture analysis to artificial intelligence: global research landscape and evolutionary trajectory of radiomics in hepatocellular carcinoma

    by Xiong Teng, Quan-Neng Luo, Yi-Di Chen, Tao Peng

    Background Hepatocellular carcinoma (HCC) poses a substantial global health burden with high morbidity and mortality rates. Radiomics, which extracts quantitative features from medical images to develop predictive models, has emerged as a promising non-invasive approach for HCC diagnosis and management. However, comprehensive analysis of research trends in this field remains limited. Methods We conducted a systematic bibliometric analysis of radiomics applications in HCC using literature from the Web of Science Core Collection (January 2006–April 2025). Publications were analyzed using CiteSpace, VOSviewer, R, and Python scripts to evaluate publication patterns, citation metrics, institutional contributions, keyword evolution, and collaboration networks. Results Among 906 included publications, we observed exponential growth, particularly accelerating after 2019. A global landscape analysis revealed China as the leader in publication volume, while the USA acted as the primary international collaboration hub. Countries like South Korea and the UK demonstrated higher average citation impact. Sun Yat-sen University was the most productive institution. Research themes evolved from fundamental texture analysis and CT/MRI applications toward predicting microvascular invasion, assessing treatment response (especially TACE), and prognostic modeling, driven recently by the deep integration of artificial intelligence (AI) and deep learning. Co-citation analysis revealed core knowledge clusters spanning radiomics methodology, clinical management, and landmark applications, demonstrating the field’s interdisciplinary nature. Conclusion Radiomics in HCC represents a rapidly expanding, AI-driven field characterized by extensive multidisciplinary collaboration. Future priorities should emphasize standardization, large-scale multicenter validation, enhanced international cooperation, and clinical translation to maximize radiomics’ potential in precision HCC oncology.


  • Toward Reliable Medical Image Segmentation by Modeling Evidential Calibrated Uncertainty

    by Ke Zou, Yidi Chen, Ling Huang, Nan Zhou, Xuedong Yuan, Xiaojing Shen, Meng Wang, Rick Siow Mong Goh, Yong Liu, Yih Chung Tham, et al.

    Medical image segmentation is critical for disease diagnosis and treatment assessment. However, concerns regarding the reliability of segmentation regions persist among clinicians, mainly attributed to the absence of confidence assessment, robustness, and calibration to accuracy. To address this, we introduce deep evidential segmentation model (DEviS), an easily implementable foundational model that seamlessly integrates into various medical image segmentation networks. DEviS not only enhances the calibration and robustness of baseline segmentation accuracy but also provides high-efficiency uncertainty estimation for reliable predictions. By leveraging SL theory, we explicitly model probability and uncertainty for medical image segmentation. Here, the Dirichlet distribution parameterizes the distribution of probabilities for different classes of the segmentation results. To generate calibrated predictions and uncertainty, we develop a trainable CUP. Furthermore, DEviS incorporates an uncertainty-aware filtering (UAF) module, which designs the metric of uncertainty-calibrated error to filter out-of-distribution (OOD) data. We conducted validation studies on publicly available datasets, including ISIC2018, KiTS2021, LiTS2017, and BraTS2019, to assess the accuracy and robustness of different backbone segmentation models enhanced by DEviS, as well as the efficiency and reliability of uncertainty estimation. Additionally, two potential clinical trials were conducted using the UAF module. The clinical application conducted on the Johns Hopkins OCT and Duke OCT-DME datasets demonstrated the effectiveness of the model in filtering OOD data. The second trial evaluated its efficacy in filtering high-quality data on the FIVES datasets. At last, the proposed DEviS method was extended to semi-supervised medical image segmentation, where it exhibited strong robustness under noisy conditions. Our code has been released in https://github.com/Cocofeat/DEviS.


  • Uncertainty-Aware Medical Diagnostic Phrase Identification and Grounding

    by Ke Zou, Yang Bai, Bo Liu, Yidi Chen, Zhihao Chen, Yang Zhou, Xuedong Yuan, Meng Wang, Xiaojing Shen, Xiaochun Cao, et al.

    Medical phrase grounding is crucial for identifying relevant regions in medical images based on phrase queries, facilitating accurate image analysis and diagnosis. However, current methods rely on manual extraction of key phrases from medical reports, reducing efficiency and increasing the workload for clinicians. Additionally, the lack of model confidence estimation limits clinical trust and usability. In this paper, we introduce a novel task—Medical Report Grounding (MRG)—which aims to directly identify diagnostic phrases and their corresponding grounding boxes from medical reports in an end-to-end manner. To address this challenge, we propose uMedGround, a a robust and reliable framework that leverages a multimodal large language model to predict diagnostic phrases by embedding a unique token, < $\mathtt {BOX}$ >, into the vocabulary to enhance detection capabilities. A vision encoder-decoder processes the embedded token and input image to generate grounding boxes. Critically, uMedGround incorporates an uncertainty-aware prediction model, significantly improving the robustness and reliability of grounding predictions. Experimental results demonstrate that uMedGround outperforms state-of-the-art medical phrase grounding methods and fine-tuned large visual-language models, validating its effectiveness and reliability. This study represents a pioneering exploration of the MRG task, marking the first-ever endeavor in this domain. Additionally, we demonstrate the applicability of uMedGround in medical visual question answering and class-based localization tasks, where it highlights visual evidence aligned with key diagnostic phrases, supporting clinicians in interpreting various types of textual inputs, including free-text reports, visual question answering queries, and class labels.


  • Open Access

    Advances in magnetic resonance imaging for the evaluation of colorectal liver metastases in context of individualized precision medicine

    by Yidi Chen, Yu Zhang, Yi Wei, Hanyu Jiang, Ling Zhang, Liling Long, Bin Song, Tao Peng