Deep Learning for Automatic Differential Diagnosis of Primary Central Nervous System Lymphoma and Glioblastoma: Multi‐Parametric Magnetic Resonance Imaging Based Convolutional Neural Network Model

2021 ◽  
Author(s):  
Wei Xia ◽  
Bin Hu ◽  
Haiqing Li ◽  
Wei Shi ◽  
Ying Tang ◽  
...  
Keyword(s):  
Neural Network ◽  
Magnetic Resonance Imaging ◽  
Central Nervous System ◽  
Nervous System ◽  
Differential Diagnosis ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Neural Network Model ◽  
Central Nervous System Lymphoma ◽  
Resonance Imaging

scholarly journals Characterization and Magnetic Resonance Imaging of a Rat Model of Human B-Cell Central Nervous System Lymphoma

2007 ◽  
Vol 13 (8) ◽  
pp. 2504-2511 ◽  
Author(s):  
Carole Soussain ◽  
Leslie L. Muldoon ◽  
Csanad Varallyay ◽  
Kristoph Jahnke ◽  
Luciana DePaula ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Central Nervous System ◽  
Nervous System ◽  
Magnetic Resonance ◽  
B Cell ◽  
Rat Model ◽  
Central Nervous System Lymphoma ◽  
Resonance Imaging ◽  
Human B Cell

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

2020 ◽  
Author(s):  
Yang Gao ◽  
Xiong Xiao ◽  
Bangcheng Han ◽  
Guilin Li ◽  
Xiaolin Ning ◽  
...  
Keyword(s):  
Neural Network ◽  
Magnetic Resonance Imaging ◽  
Differential Diagnosis ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Tumor Recurrence ◽  
Radiation Necrosis ◽  
Resonance Imaging ◽  
Glioma Recurrence ◽  
Mri Scans

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 (<i>P</i>=.02 in sensitivity and <i>P</i>&lt;.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.

scholarly journals Traditional and Advanced Neuroimaging Contributions to the Diagnosis and Differential Diagnosis of Central Nervous System Lymphoma Patients Visiting a Comprehensive Medical Center

2020 ◽  
Vol 13 (1) ◽  
pp. 6-15
Author(s):  
Ali El Dirani ◽  
Zahraa Hachem ◽  
Assaad Mohanna ◽  
Amira J. Zaylaa
Keyword(s):  
Magnetic Resonance Imaging ◽  
Computed Tomography ◽  
Central Nervous System ◽  
Nervous System ◽  
Magnetic Resonance ◽  
Imaging Techniques ◽  
Central Nervous System Lymphoma ◽  
Magnetic Resonance Images ◽  
Imaging Features ◽  
Resonance Imaging

Introduction: The diagnosis of Central Nervous System Lymphoma, especially the Primary Central Nervous System Lymphoma is carried out based on brain imaging, thus avoiding an unnecessary extend of surgery. But the traditional imaging techniques, such as Computed Tomography and Magnetic Resonance Imaging, were not satisfactory. Aims: This study was conducted to characterize the spectrum of advanced Neuroimaging, such as the advanced Magnetic Resonance Imaging features in the Central Nervous System Lymphoma patients in a comprehensive medical center in Lebanon, and compare them to what has been described in the literature review. Methods: It is a retrospective exploratory study of the clinical data and imaging features for patients admitted to the emergency and radiology departments with ages above 10 years, and who were diagnosed histopathologically with intracranial lymphoma. This study may be the first to make a Radiological evaluation of Central Nervous System Lymphoma on the local population of patients over 9 years . Results: Results showed that the study of the Computed Tomography and Magnetic Resonance Imaging data of 10 immunocompetent patients with Central Nervous System Lymphoma concurs with the previously described patient populations, except for the gender parameter. Tumors were mostly presented in the fifth or Sixth decade and they could be solitary or multi-focal. Lesions were typically located Preprint submitted to The Open Neuroimaging Journal May 14, 2020 in the supratentorial compartment. On the brain Computed Tomography, the lesions were hyperdense, and in pre-contrast Magnetic Resonance images, the lesions appeared hypointense on T1 and hyperintense on T2-Weighted images, but hypointense with respect to the grey matter. The lesions were also surrounded with a mild to moderate edema as compared to other intracranial neoplasms, such as glioblastomas. Evaluation results showed that on post-contrast Magnetic Resonance images, the majority of lesions exhibited a homogeneous enhancement of 50%. Majority of the lesions also showed a less common heterogeneous ring-like enhancement of 40%, and revealed the uncommon absence of enhancement of 10%. Calcifications, hemorrhage, and necrosis were rare findings and were present in only one patient. Conclusion: As a future prospect, studying whether the advanced imaging techniques may provide not only non-invasive and morphological characteristics but also non-invasive biological characteristics and thus accurate diagnosis could be considered.

scholarly journals Changes in cerebrospinal fluid interleukin-10 levels display better performance in predicting disease relapse than conventional magnetic resonance imaging in primary central nervous system lymphoma 

2021 ◽  
Author(s):  
Yan Zhang ◽  
Dongmei Zou ◽  
Jingjing Yin ◽  
Li Zhang ◽  
Xiao Zhang ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Central Nervous System ◽  
Cerebrospinal Fluid ◽  
Nervous System ◽  
Magnetic Resonance ◽  
Central Nervous System Lymphoma ◽  
Conventional Magnetic Resonance Imaging ◽  
Disease Relapse ◽  
Resonance Imaging

Abstract Backgroud: Establishing diagnostic and prognostic biomarkers of primary central nervous system lymphoma (PCNSL) is a challenge. This study evaluated the value of dynamic interleukin (IL)-10 cerebrospinal fluid (CSF) concentrations for prognosis and relapse prediction in PCNSL. Methods: Consecutive 40 patients newly diagnosed with PCNSL between April 2015 and April 2019 were recruited, and serial CSF specimens were collected by lumbar punctures (LP) or by Ommaya reservoir at diagnosis, treatment, and follow-up phase.Results: We confirmed that an elevated IL-10 cutoff value of 8.2 pg/mL for the diagnosis value of PCNSL showed a sensitivity of 85%. A persistent detectable CSF IL-10 level at the end of treatment was associated with poor progression-free survival (PFS) (836 vs. 481 days, p = 0.049). Within a median follow-up of 13.6 (2–55) months, 24 patients relapsed. IL-10 relapse was defined as a positive conversion in patients with undetectable IL-10 or an increased concentration compared to the last test in patients with sustained IL-10. IL-10 relapse was detected a median of 67 days (28–402 days) earlier than disease relapse in 10/16 patients. Conclusion: This study highlights a new perspective that CSF IL-10 relapse could be a surrogate marker for disease relapse and detected earlier than conventional magnetic resonance imaging (MRI) scan. Further evaluation of IL-10 monitoring in PCNSL follow-up is warranted.

Sign in / Sign up

Export Citation Format

Share Document