Feng Li

Feng Li


Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 November 2023 in Oncologie
Verified
Journal Article
Open access
PDF file
SJShijun Jia
Shijun Jia
LZLinhan Zhai
Linhan Zhai
FWFeng Wu
Feng Wu
WLWenzhi Lv
Wenzhi Lv
XMXiangde Min
Xiangde Min
FLFeng Li
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.

1 September 2020 in Pathology - Research and Practice
Verified
Journal Article
Closed access
XPXiaohong Pu
Xiaohong Pu
LZLiya Zhu
Liya Zhu
FLFeng Li
Feng Li
JZJinyu Zheng
Jinyu Zheng
HWHongyan Wu
Hongyan Wu
LQLiang Qi
Liang Qi

27 April 2024 in BMC Medical Imaging
Verified
Journal Article
Open access
PDF file
RZRui Zhang
Rui Zhang
SJShijun Jia
Shijun Jia
LZLinhan Zhai
Linhan Zhai
FWFeng Wu
Feng Wu
SZShuang Zhang
Shuang Zhang
FLFeng Li
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.

2 February 2024 in Frontiers in Medicine
Verified
Journal Article
Open access
FWFeng Wu
Feng Wu
RZRui Zhang
Rui Zhang
FLFeng Li
Feng Li
XQXiaomin Qin
Xiaomin Qin
HXHui Xing
Hui Xing
TATao Ai
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.

5 December 2025 in Frontiers in Oncology
Verified
Journal Article
Open access
PDF file
KYKui Yang
Kui Yang
WZWei Zhang
Wei Zhang
HZHao Zheng
Hao Zheng
DJDa Shuang Ji
Da Shuang Ji
HCHu Chang
Hu Chang
FLFeng Li
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.

6 January 2026 in Brain Research Bulletin
Verified
Journal Article
Open access
HLHuanhuan Liu
Huanhuan Liu
QSQian Shen
Qian Shen
XHXiaodong He
Xiaodong He
JDJiaxin Ding
Jiaxin Ding
BXBing Xiong
Bing Xiong
FLFeng Li
Feng Li
Spelling components in developmental dyslexia (DD) have been marginally reported, and there has been controversy over to what extent reading and spelling deficits share a common neuroanatomical mechanism. This study aims to explore the pattern of gray matter alteration in children with dyslexia and isolated spelling disorders (SpD). This study included 22 children with typically developing (TD), 20 children with DD, and 16 children with SpD. Voxel-based lateralization analysis was performed to calculate lateralization index (LI) and investigate the structural asymmetry of gray matter. Furthermore, the structural covariance network (SCN) was utilized to describe the changes of structural covariation network in the brain regions of interest among the DD, SpD and TD groups. The results demonstrated that compared with the TD group, the lateralization patterns of the middle frontal gyrus (MFG) and superior frontal gyrus (SFG) in the DD and SpD groups were changed. Additionally, SCN results revealed that compared with the TD group, the structural covariation of the left MFG to the Frontal_Mid_L, Occipital_Sup_R, Lingual_R, Precuneus_L and Cuneus_R were weakened in the DD group. The structural covariation between the left SFG and the Occipital_Mid_L and Occipital_Sup_L was enhanced. In contrast, the structural covariance between the left SFG and the Cuneus_R, Cingulum_Ant_R, Precuneus_L, and Precentral_L was weakened. In addition, the structural covariation of left MFG with the Supp_Motor_Area_R was increased and the structural covariation with Temporal_Sup_L was decreased in the SpD group. The structural covariation between left SFG and bilateral postcentral gyrus was increased, while that between left SFG and Parietal_Sup_L was decreased. This study displayed lateralization patterns of gray matter in children with DD and SpD, and proved that both types of dysfunctions have shared and distinct neural signatures.