Leilei Zhou

Leilei Zhou

Auto-generated profile, based on publicly available information

Leilei Zhou

Department of Medical Oncology, The Affiliated Huaian No.1 People's Hospital of Nanjing Medical University, Huai'an, 223300 Jiangsu, China

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
1 July 2021 in Radiology: Artificial Intelligence
Verified
Journal Article
Closed access
LZLeilei Zhou
Leilei Zhou
XYXindao Yin
Xindao Yin
TZTao Zhang
Tao Zhang
YFYuan Feng
Yuan Feng
YZYing Zhao
Ying Zhao
LLLingquan Lu
Lingquan Lu
A database of chest radiographs was established consisting of cardiomegaly, pneumothorax, and pleural effusion, which were used to train deep learning models for lesion detection and semiquantitati... Purpose To develop and evaluate deep learning models for the detection and semiquantitative analysis of cardiomegaly, pneumothorax, and pleural effusion on chest radiographs. Materials and Methods ...

22 January 2020 in IEEE Access
Verified
Journal Article
Open access
LZLeilei Zhou
Leilei Zhou
ZZZuoheng Zhang
Zuoheng Zhang
XYXindao Yin
Xindao Yin
HJHong-Bing Jiang
Hong-Bing Jiang
JWJie Wang
Jie Wang
JZJin-Xia Zheng
Jin-Xia Zheng
In this paper, we propose a transfer learning-based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) method for classifying fibroadenoma and invasive ductal carcinoma (IDC) in breast tumors. A total of 207 breast tumors from patients were collected and identified by pathologic diagnosis within 15 days after enhanced DCE-MRI examination; 119 patients (average age 50.52±10.33 years) had fibroadenomas, and 88 patients (average age 42.20±10.10 years) had IDCs. Two lesion-level models were built based on the InceptionV3 and VGG19 models, which were pretrained with the ImageNet dataset. The effects of different depths of transfer learning were examined. The network's performance was evaluated through five-fold cross validation. In the lesion-level models, the model based on Inception V3 obtained better results (area under the receiver operating characteristic curve (AUC) = 0.89) when the weights were frozen before layer-276. The other model based on VGG19 obtained better results (AUC = 0.87) when the weights were frozen before layer-13. Compared with the image-level models, both lesion-level models displayed better discrimination (accuracy increased by 13% and 14%) based on the VGG19 and Inception V3 models, respectively. Our research confirms that transfer learning can be utilized to classify fibroadenomas and IDCs in DCE-MRI images. Different depths of transfer learning result in different performances, and our proposed lesion-level model notably improves the classification accuracy.

7 December 2021 in Disease Markers
Verified
Journal Article
Open access
PDF file
RYRong Yao
Rong Yao
LZLeilei Zhou
Leilei Zhou
ZGZhongying Guo
Zhongying Guo
DZDahong Zhang
Dahong Zhang
TZTiecheng Zhang
Tiecheng Zhang
Objective. Precision medicine with molecular profiles has revolutionized the management of lung cancer contributing to improved prognosis. Herein, we aimed to uncover the gene expression profiling of transcription factors (TFs) in lung cancer as well as to develop a TF-based genomic model. Methods. We retrospectively curated lung cancer patients from public databases. Through comparing mRNA expression profiling between lung cancer and normal specimens, specific TFs were determined. Thereafter, a TF genomic model was developed with univariate Cox regression and stepwise multivariable Cox analyses, which was verified through the GSE72094 dataset. Gene set enrichment analyses (GSEA) were presented. Downstream targets of TFs were predicted with ChEA, JASPAR, and MotifMap projects, and their biological significance was investigated through the clusterProfiler algorithm. Results. In the TCGA cohort, we proposed a TF-based genomic model, comprised of SATB2, HLF, and NPAS2. Lung cancer individuals were remarkably stratified into high- and low-risk groups. Survival analyses uncovered that high-risk populations presented unfavorable survival outcomes. ROCs confirmed the excellent predictive potency in patients prognosis. Additionally, this model was an independent prognostic indicator in accordance with multivariate analyses. The clinical implication of the model was well verified in an independent dataset. High risk score was in relation to carcinogenic pathways. Downstream targets were characterized by immune and carcinogenic activation. Conclusion. The proposed TF genomic model acts as a promising marker for estimation of lung cancer patients outcomes. Prospective research is required for testing the clinical utility of the model in individualized management of lung cancer.