FLFeng Li

Feng Li

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Feng Li has a total of 33 co-authors. Here are the top co-authors by number of shared publications:

  • Feng Wu

    Department of Radiology, Xiangyang Central Hospital, Affiliated Hospital of Hubei University of Arts and Science, Xiangyang, China

    4 shared publications
  • Linhan Zhai

    Department of Radiology, Xiangyang Central Hospital, Affiliated Hospital of Hubei University of Arts and Science, Xiangyang, China

    3 shared publications
  • Rui Zhang

    Department of Radiology, Xiangyang Central Hospital, Affiliated Hospital of Hubei University of Arts and Science, Xiangyang, China

    3 shared publications
  • Shijun Jia

    Department of Radiology, Xiangyang Central Hospital, Affiliated Hospital of Hubei University of Arts and Science, Xiangyang, China

    3 shared publications
  • Shuang Zhang

    Department of Radiology, Xiangyang Central Hospital, Affiliated Hospital of Hubei University of Arts and Science, Xiangyang, China

    2 shared publications
  • Bing Xiong

    Department of Rehabilitation, The Second Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China (B.X.).

    1 shared publication

Latest Publications

  • Open Access

    Multiparametric quantitative MRI combining SyMRI and MUSE-DWI for noninvasive stratification of HER2 status in breast cancer

    by Kui Yang, Wei Zhang, Hao Zheng, Da Shuang Ji, Hu Chang, Feng Li

    Background: Accurate stratification of HER2 status is crucial for treatment decision-making and prognostic evaluation in breast cancer. With the recognition of HER2-low as a distinct subtype, which has recently gained clinical relevance as HER2-low patients may benefit from emerging HER2-targeted therapies, conventional pathological methods remain the gold standard; however, they are invasive and prone to sampling bias, and may not fully reflect intratumoral heterogeneity. Imaging provides a noninvasive alternative for evaluating HER2 expression. This study aimed to assess the value of synthetic MRI (SyMRI) combined with multiplexed sensitivity encoding diffusion-weighted imaging (MUSE-DWI) for noninvasive stratification of HER2 status in breast cancer. Methods: A total of 138 patients with pathologically confirmed invasive breast cancer underwent standardized MRI protocols, including SyMRI, MUSE-DWI, and DCE-MRI before biopsy or any treatment. Quantitative parameters (T 1 , T 2 , PD, ADC, and their pre-/post-contrast changes) were measured. Differences among HER2-zero, HER2-low, and HER2-overexpressing groups were analyzed. Univariate and multivariate logistic regression analyses were performed to identify independent predictors and construct nomogram models for predicting HER2 positivity and HER2-low status. Model performance was evaluated using ROC curves and calibration analysis. Results: HER2-overexpressing tumors more frequently demonstrated heterogeneous enhancement, washout-type time–intensity curves (TICs), and larger maximum diameters. In multivariable analysis, ADC, maximum diameter, T2-pre, and enhancement pattern were independent predictors of HER2 positivity (AUC = 0.940; bootstrap-corrected AUC = 0.930), whereas ADC and PD-Δ% independently predicted HER2-low status (AUC = 0.810; bootstrap-corrected AUC = 0.830). Both models showed good discrimination and calibration, and decision curve analysis indicated a favorable net clinical benefit across a wide range of threshold probabilities. Conclusions: SyMRI combined with MUSE-DWI enables noninvasive stratification of HER2 status in breast cancer. The proposed models demonstrated high diagnostic performance, good calibration, and favorable clinical utility in decision curve analysis, particularly for identifying HER2-low tumors. This imaging approach has the potential to complement biopsy and assist personalized treatment planning.


  • Open Access

    Multiparametric Quantitative MRI Combining Synthetic MRI and MUSE-DWI for Noninvasive Stratification of HER2 Expression in Breast Cancer

    by Kui Yang¹, Wei Zhang¹, Hao Zheng², Feng Li¹

    Background Accurate stratification of HER2 status is crucial for treatment decision-making and prognostic evaluation in breast cancer. With the recognition of HER2-low as a distinct subtype, conventional pathological methods remain the gold standard but are invasive and subject to sampling bias. Imaging provides a noninvasive alternative for evaluating HER2 expression. This study aimed to assess the value of synthetic MRI (SyMRI) combined with multiplexed sensitivity encoding diffusion-weighted imaging (MUSE-DWI) for noninvasive stratification of HER2 status in breast cancer. Methods A total of 138 patients with pathologically confirmed invasive breast cancer underwent standardized MRI protocols, including SyMRI, MUSE-DWI, and DCE-MRI. Quantitative parameters (T₁, T₂, PD, ADC, and their pre-/post-contrast changes) were measured. Differences among HER2-zero, HER2-low, and HER2-oe groups were compared. Univariate and multivariate logistic regression analyses were performed to identify independent predictors and construct nomogram models for predicting HER2 positivity and HER2-low. Model performance was evaluated using ROC and calibration analyses. Results HER2-over tumors more frequently demonstrated heterogeneous enhancement, washout-type TICs, and larger maximum diameters. Quantitative analysis identified ADC, maximum diameter, T2-per, and enhancement pattern as independent predictors of HER2 positivity (combined model AUC = 0.940; corrected AUC = 0.930), while ADC and PD-Δ% were independent predictors of HER2-low (combined model AUC = 0.810; corrected AUC = 0.830). Both models demonstrated good discrimination and calibration. Conclusions Multiparametric quantitative imaging with SyMRI and MUSE-DWI provides multidimensional information on HER2-related alterations in cellularity, water content, and vascular permeability. This approach offers a reliable noninvasive tool for assessing HER2 expression status, with particular clinical relevance for the identification of HER2-low and personalized treatment planning.

  • Open Access

    Predicting preoperative muscle invasion status for bladder cancer using computed tomography-based radiomics nomogram

    by Rui Zhang, Shijun Jia, Linhan Zhai, Feng Wu, Shuang Zhang, Feng Li

    Objectives: The aim of the study is to assess the efficacy of the established computed tomography (CT)-based radiomics nomogram combined with radiomics and clinical features for predicting muscle invasion status in bladder cancer (BCa). Methods: A retrospective analysis was conducted using data from patients who underwent CT urography at our institution between May 2018 and April 2023 with urothelial carcinoma of the bladder confirmed by postoperative histology. There were 196 patients enrolled in all, and each was randomized at random to either the training cohort (n = 137) or the test cohort (n = 59). Eight hundred fifty-one radiomics features in all were retrieved. For feature selection, the significance test and least absolute shrinkage and selection operator (LASSO) approaches were utilized. Subsequently, the radiomics score (Radscore) was obtained by applying linear weighting based on the selected features. The clinical and radiomics model, as well as radiomics-clinical nomogram were all established using logistic regression. Three models were evaluated using analysis of the receiver operating characteristic curve. An area under the curve (AUC) and 95% confidence intervals (CI) as well as specificity, sensitivity, accuracy, negative predictive value, and positive predictive value were included in the analysis. Radiomics-clinical nomogram’s performance was assessed based on discrimination, calibration, and clinical utility. Results: After obtaining 851 radiomics features, 12 features were ultimately selected. Histopathological grading and tortuous blood vessels were included in the clinical model. The Radscore and clinical histopathology grading were among the final predictors in the unique nomogram. The three models had an AUC of 0.811 (95% CI, 0.742–0.880), 0.845 (95% CI, 0.781–0.908), and 0.896 (95% CI, 0.846–0.947) in the training cohort and in the test cohort they were 0.808 (95% CI, 0.703–0.913), 0.847 (95% CI, 0.739–0.954), and 0.887 (95% CI, 0.803–0.971). According to the DeLong test, the radiomics-clinical nomogram’s AUC in the training cohort substantially differed from that of the clinical model (AUC: 0.896 versus 0.845, p = 0.015) and the radiomics model (AUC: 0.896 versus 0.811, p = 0.002). The Delong test in the test cohort revealed no significant difference among the three models. Conclusions: CT-based radiomics-clinical nomogram can be a useful tool for quantitatively predicting the status of muscle invasion in BCa.


  • Open Access

    Radiomics analysis based on multiparametric magnetic resonance imaging for differentiating early stage of cervical cancer

    by Feng Wu, Rui Zhang, Feng Li, Xiaomin Qin, Hui Xing, Huabing Lv, Lin Li, Tao Ai

    Objective: To investigate the performance of multiparametric magnetic resonance imaging (MRI)—based radiomics models in differentiating early stage of cervical cancer (Stage I-IIa vs. IIb-IV).Methods: One hundred patients with cervical cancer who underwent preoperative MRI between June 2020 and March 2022 were retrospectively enrolled. Training (n = 70) and testing cohorts (n = 30) were assigned by stratified random sampling. The clinical and pathological features, including age, histological subtypes, tumor grades, and node status, were compared between the two cohorts by t-test or chi-square test. Radiomics features were extracted from each volume of interest (VOI) on T2-weighted images (T2WI) and apparent diffusion coefficient (ADC) maps. The data balance of the training cohort was resampled by synthesizing minority oversampling techniques. Subsequently, the adiomics signatures were constructed by the least absolute shrinkage and selection operator algorithm and minimum-redundancy maximum-relevance with 10-fold cross-validation. Logistic regression was applied to predict the cervical cancer stages (low [I-IIa]) and (high [IIb–IV] FIGO stages). The receiver operating characteristic curve (area under the curve [AUC]) and decision curve analysis were used to assess the performance of the radiomics model.Results: The characteristics of age, histological subtypes, tumor grades, and node status were not significantly different between the low [I-IIa] and high [IIb–IV] FIGO stages (p > 0.05 for both the training and test cohorts). Three models based on T2WI, ADC maps, and the combined were developed based on six radiomics features from T2WI and three radiomics features from ADC maps, with AUCs of 0.855 (95% confidence interval [CI], 0.777–0.934) and 0.823 (95% CI, 0.727–0.919), 0.861 (95% CI, 0.785–0.936) and 0.81 (95% CI, 0.701–0.918), 0.934 (95% CI, 0.884–0.984) and 0.902 (95% CI, 0.832–0.972) in the training and test cohorts.Conclusion: The radiomics models combined T2W and ADC maps had good predictive performance in differentiating the early stage from locally advanced cervical cancer.


  • Open Access

    Predicting preoperative muscle invasion status for bladder cancer using computed tomography-based radiomics nomogram

    by Rui Zhang, Shijun Jia, Linhan Zhai, Feng Wu, Shuang Zhang, Feng Li

    Objectives The aim of the study is to assess the efficacy of the established computed tomography (CT)-based radiomics nomogram combined with radiomics and clinical features for predicting muscle invasion status in bladder cancer (BCa). Methods A retrospective analysis was conducted using data from patients who underwent CT urography at our institution between May 2018 and April 2023 with uroepithelial carcinoma of the bladder confirmed by postoperative histology. There were 196 patients enrolled in all, and each was randomized at random to either the training cohort (n = 137) or the test cohort (n = 59). 851 radiomics features in all were retrieved. For feature selection, the significance test and least absolute shrinkage and selection operator (LASSO) approaches were utilized. Subsequently, the radiomics score (Radscore) was obtained by applying linear weighting based on the selected features. The clinical and radiomics model, as well as radiomics-clinical nomogram were all established using logistic regression. Three models were evaluated using analysis of the receiver operating characteristic curve. An area under the curve (AUC) and 95% confidence intervals (CI) as well as specificity, sensitivity, accuracy, negative predictive value, and positive predictive value were included in the analysis. Radiomics-clinical nomogram's performance was assessed based on discrimination, calibration, and clinical utility. Results After obtaining 851 radiomics features, 12 features were ultimately selected. Histopathological grading and tortuous blood vessels were included in the clinical model. The Radscore and clinical histopathology grading were among the final predictors in the unique nomogram. The three models had an AUC of 0.811 (95% CI, 0.742–0.880), 0.845 (95% CI, 0.781–0.908), and 0.896 (95% CI, 0.846–0.947) in the training cohort and in the test cohort they were 0.808 (95% CI, 0.703–0.913), 0.847 (95% CI, 0.739–0.954), and 0.887 (95% CI, 0.803–0.971). According to the DeLong test, the radiomics-clinical nomogram's AUC in the training cohort substantially differed from that of the clinical model (AUC: 0.896 versus 0.845, p = 0.015) and the radiomics model (AUC: 0.896 versus 0.811, p = 0.002). The Delong test in the test cohort revealed no significant difference among the three models. The nomogram proved clinically useful, according to decision curve evaluations. Conclusions BCa muscle invasion status could be accurately predicted preoperatively by the radiomics-clinical nomogram.

  • Open Access

    Integrative machine learning algorithms for developing a consensus RNA modification-based signature for guiding clinical decision-making in bladder cancer

    by Shijun Jia, Linhan Zhai, Feng Wu, Wenzhi Lv, Xiangde Min, Shuang Zhang, Feng Li

    Objectives: Dysregulation of RNA modifications has emerged as a contributor to cancer, but the clinical implication of RNA modification-related genes remains largely unclear. The study focused on well-studied RNA modification modalities (m6A, m1A, m5C and m7G) in bladder cancer, and proposed a machine learning-based integrative approach for establishing a consensus RNA modification-based signature. Methods: Multiple publicly available bladder cancer cohorts were enrolled. A novel RNA modification-based classification was proposed via consensus clustering analysis. RNA modification-related genes were subsequently selected through WGCNA. A machine learning-based integrative framework was implemented for constructing a consensus RNA modification-based signature. Results: Most RNA modifiers were dysregulated in bladder tumours at the multi-omics levels. Two RNA modification clusters were identified, with diverse prognostic outcomes. A consensus RNA modification-based signature was established, which displayed stable and powerful efficacy in prognosis estimation. Notably, the signature was superior to conventional clinical indicators. High-risk tumours presented the activation of tumourigenic pathways, with the activation of metabolism pathways in low-risk tumours. The low-risk group was more sensitive to immune-checkpoint blockade, with the higher sensitivity of the high-risk group to cisplatin and paclitaxel. Genes in the signature: AKR1B1, ANXA1, CCNL2, OAS1, PTPN6, SPINK1 and TNFRSF14 were specially expressed in distinct T lymphocytes of bladder tumours at the single-cell level, potentially participating in T cell-mediated antitumour immunity. They were transcriptionally and post-transcriptionally modulated, and might become potentially actionable therapeutic targets. Conclusions: Altogether, the consensus RNA modification-based signature may act as a reliable and hopeful tool for improving clinical decision-making for individual bladder cancer patients.


  • Target molecular treatment markers in Intrahepatic Cholangiocarcinoma based on Chinese population

    by Xiaohong Pu, Liya Zhu, Feng Li, Jinyu Zheng, Hongyan Wu, Yao Fu, Jun Chen, Liang Qi