Machine learning applications for the differentiation of primary central nervous system lymphoma from glioblastoma on imaging: a systematic review and meta-analysis

Abstract

<jats:sec><jats:title>OBJECTIVE</jats:title><jats:p>Glioblastoma (GBM) and primary central nervous system lymphoma (PCNSL) are common intracranial pathologies encountered by neurosurgeons. They often may have similar radiological findings, making diagnosis difficult without surgical biopsy; however, management is quite different between these two entities. Recently, predictive analytics, including machine learning (ML), have garnered attention for their potential to aid in the diagnostic assessment of a variety of pathologies. Several ML algorithms have recently been designed to differentiate GBM from PCNSL radiologically with a high sensitivity and specificity. The objective of this systematic review and meta-analysis was to evaluate the implementation of ML algorithms in differentiating GBM and PCNSL.</jats:p></jats:sec><jats:sec><jats:title>METHODS</jats:title><jats:p>The authors performed a systematic review of the literature using PubMed in accordance with PRISMA guidelines to select and evaluate studies that included themes of ML and brain tumors. These studies were further narrowed down to focus on works published between January 2008 and May 2018 addressing the use of ML in training models to distinguish between GBM and PCNSL on radiological imaging. Outcomes assessed were test characteristics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).</jats:p></jats:sec><jats:sec><jats:title>RESULTS</jats:title><jats:p>Eight studies were identified addressing use of ML in training classifiers to distinguish between GBM and PCNSL on radiological imaging. ML performed well with the lowest reported AUC being 0.878. In studies in which ML was directly compared with radiologists, ML performed better than or as well as the radiologists. However, when ML was applied to an external data set, it performed more poorly.</jats:p></jats:sec><jats:sec><jats:title>CONCLUSIONS</jats:title><jats:p>Few studies have applied ML to solve the problem of differentiating GBM from PCNSL using imaging alone. Of the currently published studies, ML algorithms have demonstrated promising results and certainly have the potential to aid radiologists with difficult cases, which could expedite the neurosurgical decision-making process. It is likely that ML algorithms will help to optimize neurosurgical patient outcomes as well as the cost-effectiveness of neurosurgical care if the problem of overfitting can be overcome.</jats:p></jats:sec>

Radiomics-Based Machine Learning in Differentiation Between Glioblastoma and Metastatic Brain Tumors

“…Given that MR scan is the conventional radiological examination for patients, TA on T1C has the potential to serve as a feasible solution in clinical application without requiring additional fees. Previous studies have illustrated that TA combined with machine learning could assist in the diagnosis of various brain tumors, such as GBM from primary central nerve system lymphoma and meningioma from GBM (16, 17). Moreover, it has also been applied in tumor grade system and gene mutation prediction (18–22).…”

Multiparametric‐MRI‐Based Radiomics Model for Differentiating Primary Central Nervous System Lymphoma From Glioblastoma: Development and Cross‐Vendor Validation

“…LASSO is a generalized linear model (GLM) that performs both feature selection and regularization to enhance the classification accuracy and interpretability of the model, 28 and has shown advantages over other classifiers in radiomics studies. 13,18 For each MRI sequence, a single-sequence radiomics model was trained using 10-fold cross-validation. Multisequence radiomics models were generated by integrating single-sequence radiomics models using multivariable logistic regression with all possible combinations of sequences.…”

“…9,[14][15][16][17] This has resulted in concerns about the risk of overfitting the models that may be biased by subtle differences between MRI hardware or imaging parameters. The generalizability of radiomics models to correctly interpret data acquired by different MR scanners with different protocol parameters is important 18 and would allow the application of radiomics in clinical practice. Furthermore, while in previous studies the performance of radiomics models were evaluated against radiologists, 9,14,16,17 models were initially developed as support tools to assist radiologists rather than to replace them.…”

Comparison of Multivariable Logistic Regression and Other Machine Learning Algorithms for Prognostic Prediction Studies in Pregnancy Care: Systematic Review and Meta-Analysis

“…Beyond the algorithm issue, the development of machine learning models also requires an adequate methodology and interpretable results [15]. Biased conclusions should be avoided when describing machine learning predictive performances [11,16]. Standard guidelines are important when investigating and reviewing machine learning applications in clinical prediction modeling [15,17].…”