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


Fuli Zhang has a total of 43 co-authors. Here are the top co-authors by number of shared publications:

  • Diandian Chen

    Radiation Oncology Department, The Seventh Medical Center of Chinese People's Liberation Army PLA General Hospital, Beijing, China

    5 shared publications
  • Huayong Jiang

    Radiation Oncology Department, The Seventh Medical Center of Chinese People's Liberation Army PLA General Hospital, Beijing, China

    5 shared publications
  • Na Lu

    Radiation Oncology Department, The Seventh Medical Center of Chinese People's Liberation Army PLA General Hospital, Beijing, China

    5 shared publications
  • Yadi Wang

    Radiation Oncology Department, The Seventh Medical Center of Chinese People's Liberation Army (PLA) General Hospital, Beijing, China

    5 shared publications
  • Mingmin Zheng

    Radiation Oncology Department, the Military General Hospital of Beijing PLA, Dongcheng, Beijing, China

    4 shared publications
  • Qiusheng Wang

    School of Automation Science and Electrical Engineering, Beihang University, Beijing, China

    4 shared publications

Latest Publications

  • Open Access

    A hybrid quality assurance framework for total marrow and lymphoid irradiation: Integrating independent 3D dose calculation and measurement-based verification

    by Qin Wang, Yanjun Yu, Wenjie Ni, Linan Song, Hui Yang, Xiunan Wang, Xiaofeng Mu, Fuli Zhang

  • Open Access

    Integration of intratumoral/peritumoral radiomics and deep learning for predicting overall survival in non-small cell lung cancer patients: a multicenter study

    by Yongxin Liu, Yuteng Pan, Qiusheng Wang, Huayong Jiang, Na Lu, Diandian Chen, Yanjun Yu, Yanxiang Gao, Huijuan Zhang, Yinglun Sun, et al.

    Background: Prognostic assessment of non-small cell lung cancer (NSCLC) relies on TNM staging, yet tumor heterogeneity limits its accuracy. This study aimed to develop a model for improving the prediction of overall survival (OS) in NSCLC patients receiving radiotherapy, which integrated intratumoral/peritumoral radiomics features and 3D deep learning (DL) features. Methods: A total of 303 NSCLC patients from three centers were retrospectively enrolled. Radiomics features were extracted from intratumoral and 3/6/9 mm peritumoral regions on CT scans. A network named 3D-SE-ResNet was proposed to extract DL features. Lasso-Cox and principal component analysis (PCA) were used to integrate multidimensional features to establish a combined model. Performance was evaluated via the concordance index (C-index) and area under the curve (AUC). Survival differences were visualized through Kaplan–Meier curves. Results: The 6 mm expansion peritumoral radiomics features analysis showed the best performance (C-index: 0.63). The DL features outperformed the radiomics features (C-index: 0.74 vs 0.63). The combined model achieved the highest accuracy (C-index: 0.77/0.73 across datasets). K–M analysis confirmed significant survival differences (log-rank P < 0.001). Conclusion: The combined model integrates intratumoral/peritumoral radiomics features and 3D DL features and effectively predicts the OS of NSCLC patients, offering a novel tool for personalized radiotherapy strategies.


  • Open Access

    Applying a novel two-step deep learning network to improve the automatic delineation of esophagus in non-small cell lung cancer radiotherapy

    by Fuli Zhang, Qiusheng Wang, Na Lu, Diandian Chen, Huayong Jiang, Anning Yang, Yanjun Yu, Yadi Wang

    Purpose: To introduce a model for automatic segmentation of thoracic organs at risk (OARs), especially the esophagus, in non-small cell lung cancer radiotherapy, using a novel two-step deep learning network.Materials and methods: A total of 59 lung cancer patients’ CT images were enrolled, of which 39 patients were randomly selected as the training set, 8 patients as the validation set, and 12 patients as the testing set. The automatic segmentations of the six OARs including the esophagus were carried out. In addition, two sets of treatment plans were made on the basis of the manually delineated tumor and OARs (Plan1) as well as the manually delineated tumor and the automatically delineated OARs (Plan2). The Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and average surface distance (ASD) of the proposed model were compared with those of U-Net as a benchmark. Next, two groups of plans were also compared according to the dose–volume histogram parameters.Results: The DSC, HD95, and ASD of the proposed model were better than those of U-Net, while the two groups of plans were almost the same. The highest mean DSC of the proposed method was 0.94 for the left lung, and the lowest HD95 and ASD were 3.78 and 1.16 mm for the trachea, respectively. Moreover, the DSC reached 0.73 for the esophagus.Conclusions: The two-step segmentation method can accurately segment the OARs of lung cancer. The mean DSC of the esophagus realized preliminary clinical significance (>0.70). Choosing different deep learning networks based on different characteristics of organs offers a new option for automatic segmentation in radiotherapy.


  • Automatic segmentation of the tumor in nonsmall‐cell lung cancer by combining coarse and fine segmentation

    by Fuli Zhang, Qiusheng Wang, Enyu Fan, Na Lu, Diandian Chen, Huayong Jiang, Yadi Wang

    ObjectivesRadiotherapy plays an important role in the treatment of nonsmall-cell lung cancer (NSCLC). Accurate delineation of tumor is the key to successful radiotherapy. Compared with the commonly used manual delineation ways, which are time-consuming and laborious, the automatic segmentation methods based on deep learning can greatly improve the treatment efficiency. MethodsIn this paper, we introduce an automatic segmentation method by combining coarse and fine segmentations for NSCLC. Coarse segmentation network is the first level, identifing the rough region of the tumor. In this network, according to the tissue structure distribution of the thoracic cavity where tumor is located, we designed a competition method between tumors and organs at risk (OARs), which can increase the proportion of the identified tumor covering the ground truth and reduce false identification. Fine segmentation network is the second level, carrying out precise segmentation on the results of the coarse level. These two networks are independent of each other during training. When they are used, morphological processing of small scale corrosion and large scale expansion is used for the coarse segmentation results, and the outcomes are sent to the fine segmentation part as input, so as to achieve the complementary advantages of the two networks. ResultsIn the experiment, CT images of 200 patients with NSCLC are used to train the network, and CT images of 60 patients are used to test. Finally, our method produced the Dice similarity coefficient of 0.78 +/- 0.10. ConclusionsThe experimental results show that the proposed method can accurately segment the tumor with NSCLC, and can also provide support for clinical diagnosis and treatment.


  • Open Access

    Geometric and Dosimetric Evaluation of the Automatic Delineation of Organs at Risk (OARs) in Non-Small-Cell Lung Cancer Radiotherapy Based on a Modified DenseNet Deep Learning Network

    by Fuli Zhang, Qiusheng Wang, Anning Yang, Na Lu, Huayong Jiang, Diandian Chen, Yanjun Yu, Yadi Wang

    Purpose: To introduce an end-to-end automatic segmentation model for organs at risk (OARs) in thoracic CT images based on modified DenseNet, and reduce the workload of radiation oncologists.Materials and Methods: The computed tomography (CT) images of 36 lung cancer patients were included in this study, of which 27 patients’ images were randomly selected as the training set, 9 patients’ as the testing set. The validation set was generated by cross validation and 6 patients’ images were randomly selected from the training set during each epoch as the validation set. The autosegmentation task of the left and right lungs, spinal cord, heart, trachea and esophagus was implemented, and the whole training time was approximately 5 hours. Geometric evaluation metrics including the Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95) and average surface distance (ASD), were used to assess the autosegmentation performance of OARs based on the proposed model and were compared with those based on U-Net as benchmarks. Then, two sets of treatment plans were optimized based on the manually contoured targets and OARs (Plan1), as well as the manually contours targets and the automatically contoured OARs (Plan2). Dosimetric parameters, including Dmax, Dmean and Vx, of OARs were obtained and compared.Results: The DSC, HD95 and ASD of the proposed model were better than those of U-Net. The differences in the DSC of the spinal cord and esophagus, differences in the HD95 of the spinal cord, heart, trachea and esophagus, as well as differences in the ASD of the spinal cord were statistically significant between the two models (P0.05). Moreover, compared with manual segmentation, autosegmentation significantly reduced the contouring time by nearly 40.7% (P<0.05).Conclusions: The bilateral lungs, spinal cord, heart and trachea could be accurately delineated using the proposed model in this study; however, the automatic segmentation effect of the esophagus must still be further improved. The concept of feature map reuse provides a new idea for automatic medical image segmentation.


  • Open Access

    Automatic Segmentation of the Gross Target Volume in Non-Small Cell Lung Cancer Using a Modified Version of ResNet

    by Fuli Zhang, Qiusheng Wang, Haipeng Li

    Radiotherapy plays an important role in the treatment of non-small cell lung cancer. Accurate segmentation of the gross target volume is very important for successful radiotherapy delivery. Deep learning techniques can obtain fast and accurate segmentation, which is independent of experts’ experience and saves time compared with manual delineation. In this paper, we introduce a modified version of ResNet and apply it to segment the gross target volume in computed tomography images of patients with non-small cell lung cancer. Normalization was applied to reduce the differences among images and data augmentation techniques were employed to further enrich the data of the training set. Two different residual convolutional blocks were used to efficiently extract the deep features of the computed tomography images, and the features from all levels of the ResNet were merged into a single output. This simple design achieved a fusion of deep semantic features and shallow appearance features to generate dense pixel outputs. The test loss tended to be stable after 50 training epochs, and the segmentation took 21 ms per computed tomography image. The average evaluation metrics were: Dice similarity coefficient, 0.73; Jaccard similarity coefficient, 0.68; true positive rate, 0.71; and false positive rate, 0.0012. Those results were better than those of U-Net, which was used as a benchmark. The modified ResNet directly extracted multi-scale context features from original input images. Thus, the proposed automatic segmentation method can quickly segment the gross target volume in non-small cell lung cancer cases and be applied to improve consistency in contouring.


  • Dosimetric evaluation of VMAT radiation therapy technique for breast cancer after conservative surgery based on three different types of multileaf collimators

    by Fuli Zhang, Weidong Xu, Huayong Jiang, Yadi Wang, Junmao Gao, Qingzhi Liu

    Objective: Radiotherapy combined with conservative surgery plays an important role in the treatment of early-stage breast cancer. Volumetric modulated arc therapy (VMAT) has been introduced into clinical practice. The purpose of this study was to investigate the dosimetric effects of different multileaf collimators (MLC) on VMAT radiotherapy plans for treating breast cancer. Methods: Fifteen breast cancer patients who were treated using a conventional technique in our department were selected to participate in this retrospective analysis. VMAT plans based on three types of Elekta MLCs [Beam Modulator (BM) with 4-mm leaf width, Agility with 5-mm leaf width and MLCi2 with 10-mm leaf width] were independently generated for each patient. Plan comparisons were performed based on dose-volume histogram (DVH) analysis including dosimetric parameters such as the homogeneity index (HI), conformity index (CI), Dmax, Dmin, and Dmean for the planning treatment volume (PTV), in addition to dose-volume parameters for the organs at risk (OARs). The delivery efficiency of the three types of MLCs was compared in terms of the beam delivery time and the monitor units (MUs) per fraction for each plan. Results: Both target uniformity and conformity were improved in plans for Agility and BM MLC compared with the plan using MLCi2. The mean HI decreased from 1.14 for MLCi2 to 1.13 for BM and 1.10 for Agility, while the mean CI increased from 0.68 for MLCi2 to 0.73 for BM and 0.75 for Agility. Furthermore, at both low and high dose levels, smaller volumes of ipsilateral lung, heart, contralateral lung, and breast were irradiated with Agility MLC than with the other two types of MLCs. The delivery time with Agility MLC was reduced by 10.8% and 32.1%, respectively, compared with that for MLCi2 and BM. Conclusion: Our results indicate that the Agility MLC exhibits a dosimetric advantage and a significant improvement in delivery efficiency for the treatment of breast cancer using VMAT.

  • Open Access

    The Expression and Clinical Outcome of pCHK2-Thr68 and pCDC25C-Ser216 in Breast Cancer

    by Huayong Jiang, Bin Wang, Fuli Zhang, Yuanyu Qian, Chia-Chen Chuang, Mingzhen Ying, Yajie Wang, Li Zuo

    Checkpoint kinase 2 (CHK2) and cell division cycle 25C (CDC25C) are two proteins involved in the DNA damage response pathway, playing essential roles in maintaining genome integrity. As one of the major hallmarks of abnormal cellular division, genomic instability occurs in most cancers. In this study, we identified the functional expression of pCHK2-Thr68 and pCDC25C-Ser216 in breast cancer, as well as its association with breast cancer survival. Tissue microarray analysis using immunohistochemistry was constructed to identify the expression of pCHK2-Thr68 and pCDC25C-Ser216 in 292 female breast cancer patients. The relationship among protein expression, clinicopathological factors (e.g., human epidermal growth factor receptor 2 (HER 2), tumor size, tumor-node-metastasis (TNM) classification), and overall survival of the breast cancer tissues were analyzed using Pearson’s χ-square (χ2) test, Fisher’s exact test, multivariate logistic regression and Kaplan–Meier survival analysis. Significantly higher expressions of pCHK2-Thr68 and pCDC25C-Ser216 were observed in the nucleus of the breast cancer cells compared to the paracancerous tissue (pCHK2-Thr68, 20.38% vs. 0%; pCDC25C-Ser216, 82.26% vs. 24.24%). The expression of pCHK2-Thr68 and pCDC25C-Ser216 in breast cancer showed a positive linear correlation (p = 0.026). High expression of pCHK2-Thr68 was associated with decreased patient survival (p = 0.001), but was not an independent prognostic factor. Our results suggest that pCHK2-Thr68 and pCDC25C-Ser216 play important roles in breast cancer and may be potential treatment targets.


  • A dosimetric evaluation of flattening filter-free volumetric modulated arc therapy for postoperative treatment of cervical cancer ⁎

    by Fuli Zhang, Huayong Jiang, Weidong Xu, Yadi Wang, Junmao Gao, Qingzhi Liu, Ping Wang, Na Lu, Diandian Chen, Bo Yao, et al.

    Objective: The aim of the study was to compare flattening filter-free (FFF) beams and conventional flattening filter (FF) beams in volumetric modulated arc therapy (VMAT) for cervical cancer after surgery, through a retrospective planning study. Methods: VMAT plans of FFF beams and normal FF beams were designed for a cohort of 15 patients. The prescribed dose was 45 Gy to 1.8 Gy per fraction, and at least 95% of the planning target volume received this dose. Doses were computed with a commercially available treatment planning system using a Monte Carlo (MC) algorithm. Plans were compared according to dose-volume histogram analysis in terms of planning target volume homogeneity and conformity indices (HI and CI), as well as organs at risk (OAR) dose and volume parameters. Results: FFF-VMAT was similar to FF-VMAT in terms of CI, but inferior to FF-VMAT considering HI. No statistically differences were observed between FFF-VMAT and FF-VMAT in following organ at risks including pelvic bone marrow, small bowel, bladder, rectum, and normal tissue (NT). . Conclusion: For patients with cervical cancer after hysterectomy, the FFF beam achieved target and OAR dose distribution similar to that of the FF beam. Reduction of beam-on time in cervical cancer is beneficial.