GLGuilin Li

Guilin 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


Guilin Li has a total of 34 co-authors. Here are the top co-authors by number of shared publications:

  • Yanjiao He

    Department of Neuropathology, Beijing Neurosurgical Institute, Beijing Tiantan Hospital, Capital Medical University, China National Clinical Research Center for Neurological Diseases (NCRC-ND), Brain Tumor Center, Beijing Institute for Brain Disorders, Beijing Key Brain Tumor Laboratory

    3 shared publications
  • Antonio Di Ieva

    Computational NeuroSurgery (CNS) Lab, Macquarie Medical School, Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, Australia

    2 shared publications
  • Bangcheng Han

    School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing 100191, China

    2 shared publications
  • Defeng Wang

    School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing, China

    2 shared publications
  • Jiang Du

    Department of Neuropathology, Beijing Neurosurgical Institute, Beijing Tiantan Hospital, Capital Medical University, China National Clinical Research Center for Neurological Diseases (NCRC-ND), Center of Brain Tumor, Beijing Institute for Brain Disorders, Beijing Key Laboratory of Brain Tumor, Beijing, 100050, China

    2 shared publications
  • Jingyi Fang

    Department of Neuropathology, Beijing Neurosurgical Institute, Beijing Tiantan Hospital, Capital Medical University, China National Clinical Research Center for Neurological Diseases (NCRC-ND), Center of Brain Tumor, Beijing Institute for Brain Disorders, Beijing Key Laboratory of Brain Tumor, Beijing, 100050, China

    2 shared publications

Latest Publications

  • Open Access

    Deep Learning Methodology for Differentiating Glioma Recurrence From Radiation Necrosis Using Multimodal Magnetic Resonance Imaging: Algorithm Development and Validation

    by Yang Gao, Xiong Xiao, Bangcheng Han, Guilin Li, Xiaolin Ning, Defeng Wang, Weidong Cai, Ron Kikinis, Shlomo Berkovsky, Antonio Di Ieva, et al.

    Background: The radiological differential diagnosis between tumor recurrence and radiation-induced necrosis (ie, pseudoprogression) is of paramount importance in the management of glioma patients.Objective: This research aims to develop a deep learning methodology for automated differentiation of tumor recurrence from radiation necrosis based on routine magnetic resonance imaging (MRI) scans.Methods: In this retrospective study, 146 patients who underwent radiation therapy after glioma resection and presented with suspected recurrent lesions at the follow-up MRI examination were selected for analysis. Routine MRI scans were acquired from each patient, including T1, T2, and gadolinium-contrast-enhanced T1 sequences. Of those cases, 96 (65.8%) were confirmed as glioma recurrence on postsurgical pathological examination, while 50 (34.2%) were diagnosed as necrosis. A light-weighted deep neural network (DNN) (ie, efficient radionecrosis neural network [ERN-Net]) was proposed to learn radiological features of gliomas and necrosis from MRI scans. Sensitivity, specificity, accuracy, and area under the curve (AUC) were used to evaluate performance of the model in both image-wise and subject-wise classifications. Preoperative diagnostic performance of the model was also compared to that of the state-of-the-art DNN models and five experienced neurosurgeons.Results: DNN models based on multimodal MRI outperformed single-modal models. ERN-Net achieved the highest AUC in both image-wise (0.915) and subject-wise (0.958) classification tasks. The evaluated DNN models achieved an average sensitivity of 0.947 (SD 0.033), specificity of 0.817 (SD 0.075), and accuracy of 0.903 (SD 0.026), which were significantly better than the tested neurosurgeons (P=.02 in sensitivity and P<.001 in specificity and accuracy).Conclusions: Deep learning offers a useful computational tool for the differential diagnosis between recurrent gliomas and necrosis. The proposed ERN-Net model, a simple and effective DNN model, achieved excellent performance on routine MRI scans and showed a high clinical applicability.


  • Open Access

    Deep Learning Methodology for Differentiating Glioma Recurrence From Radiation Necrosis Using Multimodal Magnetic Resonance Imaging: Algorithm Development and Validation (Preprint)

    by Yang Gao, Xiong Xiao, Bangcheng Han, Guilin Li, Xiaolin Ning, Defeng Wang, Weidong Cai, Ron Kikinis, Shlomo Berkovsky, Antonio Di Ieva, et al.

    BACKGROUND The radiological differential diagnosis between tumor recurrence and radiation-induced necrosis (ie, pseudoprogression) is of paramount importance in the management of glioma patients. OBJECTIVE This research aims to develop a deep learning methodology for automated differentiation of tumor recurrence from radiation necrosis based on routine magnetic resonance imaging (MRI) scans. METHODS In this retrospective study, 146 patients who underwent radiation therapy after glioma resection and presented with suspected recurrent lesions at the follow-up MRI examination were selected for analysis. Routine MRI scans were acquired from each patient, including T1, T2, and gadolinium-contrast-enhanced T1 sequences. Of those cases, 96 (65.8%) were confirmed as glioma recurrence on postsurgical pathological examination, while 50 (34.2%) were diagnosed as necrosis. A light-weighted deep neural network (DNN) (ie, efficient radionecrosis neural network [ERN-Net]) was proposed to learn radiological features of gliomas and necrosis from MRI scans. Sensitivity, specificity, accuracy, and area under the curve (AUC) were used to evaluate performance of the model in both image-wise and subject-wise classifications. Preoperative diagnostic performance of the model was also compared to that of the state-of-the-art DNN models and five experienced neurosurgeons. RESULTS DNN models based on multimodal MRI outperformed single-modal models. ERN-Net achieved the highest AUC in both image-wise (0.915) and subject-wise (0.958) classification tasks. The evaluated DNN models achieved an average sensitivity of 0.947 (SD 0.033), specificity of 0.817 (SD 0.075), and accuracy of 0.903 (SD 0.026), which were significantly better than the tested neurosurgeons (P=.02 in sensitivity and P<.001 in specificity and accuracy). CONCLUSIONS Deep learning offers a useful computational tool for the differential diagnosis between recurrent gliomas and necrosis. The proposed ERN-Net model, a simple and effective DNN model, achieved excellent performance on routine MRI scans and showed a high clinical applicability.


  • Evaluation of EZH2 expression, BRAF V600E mutation, and CDKN2A/B deletions in epithelioid glioblastoma and anaplastic pleomorphic xanthoastrocytoma

    by Junmei Wang, Zhaoxia Liu, Yun Cui, Yuqing Liu, Jingyi Fang, Li Xu, Yanjiao He, Jiang Du, Yujin Su, Wanjing Zou, et al.

    Epithelioid glioblastoma (EGBM) and anaplastic pleomorphic xanthoastrocytoma (APXA) are two rare entities with different prognoses. However, they share certain morphological and molecular features.


  • Open Access

    Pituitary Adenoma Associated With Rathke’s Cleft Cyst: Report of 15 Cases

    by Wenhao Wu, Guijun Jia, Wang Jia, Guilin Li, Junting Zhang, Liwei Zhang

    Objective:The concomitant presence of pituitary adenoma (PA) with Rathke’s cleft cyst (RCC) is rare, and most of the literature published is case reports. RCC’s clinicopathological features have not been well described.Methods:We retrospectively reviewed the data of 15 patients with PA associated with RCC. We also provide a systematic review of the literature.Results:The patients included males and seven females. The transsphenoidal approach was adopted for all patients except one, who underwent right frontal craniotomy. Complete resection was achieved in nine cases, with grossly complete resection in five and partial removal in one. Of the PA subtypes available, five patients had nonfunctional pituitary adenomas, three had multiple-hormone secreting PAs, three had prolactin-producing adenomas, and one had a growth hormone-secreting adenoma. All of the patients’ symptoms and abnormal endocrine test results resolved or were relieved after surgery. The mean follow-up time was 27.6±16.6 months. All of the patients recovered well and had no signs of recurrence during their follow-up, except for one patient.Conclusions:A cyst-like signal indicated on images or cystic materials encountered during operation in a patient with PA suggest the possibility of PA associated with RCC. Transsphenoidal surgery is the preferred approach and can resolve or relieve patients’ symptoms to a great extent. Histologically, nonfunctional PA and prolactinoma are the most common types of PA coexisting with RCC; the mechanisms of PA associated with RCC need to be investigated further.


  • The clinicopathological features of liponeurocytoma

    by Li Xu, Jiang Du, Junmei Wang, Jingyi Fang, Zhaoxia Liu, Yanjiao He, Guilin Li

    To discuss the clinicopathological features of liponeurocytoma, we retrospectively reviewed three liponeurocytoma cases and compared their immunophenotypes and genotypes with those of similar tumors. Furthermore, we reviewed the literature and compared the similarities and differences between cerebellar and intraventricular liponeurocytomas. Two cerebellar and one intraventricular liponeurocytomas were included in the present study. The liponeurocytomas comprised small tumor cells and lipomatous cells. The tumor cells expressed SYN, MAP-2, and NeuN. One case showed atypical histological features. By reviewing the literature, we found that cerebellar liponeurocytoma tended to be more common in females, whereas the converse was true for intraventricular liponeurocytoma. Compared with cerebellar liponeurocytoma, intraventricular liponeurocytoma was more commonly noted in younger adult patients. A high MIB-1 index (>10%) and incomplete tumor resection might represent adverse prognostic factors in patients with liponeurocytoma. We suggest that ‘central liponeurocytoma’ should be used to include all putative liponeurocytoma sites. The present study identified several morphological, immunohistochemical, and genetic features that may aid in the differential diagnosis of liponeurocytoma. In addition, surgery should be the preferred treatment, and complete tumor resection should be the goal. Additional cases with long-term follow-up are needed to develop optimal management protocols for liponeurocytoma.


  • A clinicopathologic study of extraventricular neurocytoma

    by Li Xu, Zhaolian Ouyang, Junmei Wang, Zhaoxia Liu, Jingyi Fang, Jiang Du, Yanjiao He, Guilin Li

    In 2007, extraventricular neurocytoma was classified as a separate entity among glioneuronal tumors. However, extraventricular neurocytoma is not fully understood and may be misdiagnosed. Here, we describe the clinical and pathological features, prognoses, and treatments of 13 extraventricular neurocytoma cases, and compare their immunophenotypes with those of oligodendroglioma, diffuse astrocytoma, and ependymoma. Six typical and 7 atypical cases comprised the 13 extraventricular neurocytoma cases. Histological features included oligodendroglioma-like perinuclear halo, neuropil-like matrix, ganglion or ganglioid cells, perivascular pseudorosettes, vessel hyalinization, calcifications, and myxoid degeneration. Atypical histological features included increased mitotic figures, focal necrosis, endothelial cell proliferation, and/or a Ki-67 index of >2%. All lesions expressed synaptophysin and microtubule-associated protein-2, which distinguished them from other similar tumors. Two patients with atypical extraventricular neurocytoma had tumor recurrence, one of whom had cerebrospinal fluid dissemination, suggesting that atypical histological features might represent adverse prognostic factors. In conclusion, the present study identified morphological and immunohistochemical features that would aid the differential diagnosis of extraventricular neurocytoma. In addition, radiotherapy with subtotal resection could be considered an effective treatment for extraventricular neurocytoma, but because a pediatric patient died of intracranial hemorrhage during radiotherapy, radiotherapy-related side effects should be considered, especially when treating children. Additional cases with long-term follow-up are needed to develop optimal management protocols for extraventricular neurocytoma.