Machine learning for semi­automated classification of glioblastoma, brain metastasis and central nervous system lymphoma using magnetic resonance advanced imaging

Objectives To evaluate the diagnostic performance of multiple machine learning classifier models derived from first-order histogram texture parameters extracted from T1-weighted contrast-enhanced images in differentiating glioblastoma and primary central nervous system lymphoma. Methods Retrospective study with 97 glioblastoma and 46 primary central nervous system lymphoma patients. Thirty-six different combinations of classifier models and feature selection techniques were evaluated. Five-fold nested cross-validation was performed. Model performance was assessed for whole tumour and largest single slice using receiver operating characteristic curve. Results The cross-validated model performance was relatively similar for the top performing models for both whole tumour and largest single slice (area under the curve 0.909–0.924). However, there was a considerable difference between the worst performing model (logistic regression with full feature set, area under the curve 0.737) and the highest performing model for whole tumour (least absolute shrinkage and selection operator model with correlation filter, area under the curve 0.924). For single slice, the multilayer perceptron model with correlation filter had the highest performance (area under the curve 0.914). No significant difference was seen between the diagnostic performance of the top performing model for both whole tumour and largest single slice. Conclusions T1 contrast-enhanced derived first-order texture analysis can differentiate between glioblastoma and primary central nervous system lymphoma with good diagnostic performance. The machine learning performance can vary significantly depending on the model and feature selection methods. Largest single slice and whole tumour analysis show comparable diagnostic performance.

Download Full-text

Central Nervous System Lymphoma: Characteristic Findings on Traditional and Advanced Imaging

American Journal of Neuroradiology ◽

10.3174/ajnr.a2171 ◽

2010 ◽

Vol 32 (6) ◽

pp. 984-992 ◽

Cited By ~ 183

Author(s):

I.S. Haldorsen ◽

A. Espeland ◽

E.-M. Larsson

Keyword(s):

Central Nervous System ◽

Nervous System ◽

Central Nervous System Lymphoma ◽

Advanced Imaging

Download Full-text

A case of primary central nervous system lymphoma mimic neuromyelitis optica

Translational Neuroscience ◽

10.1515/tnsci-2020-0005 ◽

2020 ◽

Vol 11 (1) ◽

pp. 28-33

Author(s):

Xixi Sheng ◽

Mingwei Xu ◽

Xia Li

Keyword(s):

Central Nervous System ◽

Nervous System ◽

Magnetic Resonance ◽

Neuromyelitis Optica ◽

Central Nervous System Lymphoma ◽

B Cell Lymphoma ◽

Clinical Findings ◽

Resonance Spectroscopy ◽

Steroid Pulse ◽

Positron Emission

AbstractPrimary central nervous system lymphoma (PCNSL) is rare. And the symptoms of PCNSL are atypical, it is extremely easy to be misdiagnosed as other diseases. However, early treatment is crucial which is requesting early diagnosis. We report a case of a 47-year-old man who was initially diagnosed as neuromyelitis optica (NMO) on the basis of clinical findings, slightly high Aquaporin4 (AQP4) (1:10) and high signals of magnetic resonance imaging. Though his symptoms progressively improved after steroid pulse treatment, but worse when steroid was decreased to 40 mg per day. We considered the patient should be diagnosed as PCNSL. After the examination of magnetic resonance spectroscopy (MRS) and positron emission tomography (PET), the results indicated PCNSL was most possible. Therefore we gave him stereotactic biopsy of deep of supratentorial, which showed non-Hodgkin malignant B-cell lymphoma.

Download Full-text