Differentiating glioblastoma multiforme from cerebral lymphoma: application of advanced texture analysis of quantitative apparent diffusion coefficients

Deep learning for image classification between primary central nervous system lymphoma and glioblastoma in corpus callosal tumors

Objectives

It can be challenging in some situations to distinguish primary central nervous system lymphoma (PCNSL) from glioblastoma (GBM) based on magnetic resonance imaging (MRI) scans, especially those involving the corpus callosum. The objective of this study was to assess the diagnostic performance of deep learning (DL) models between PCNSLs and GBMs in corpus callosal tumors.

Materials and Methods

The axial T1-weighted gadolinium-enhanced MRI scans of 274 individuals with pathologically confirmed PCNSL (n = 94) and GBM (n = 180) were examined. After image pooling, pre-operative MRI scans were randomly split with an 80/20 procedure into a training dataset (n = 709) and a testing dataset (n = 177) for DL model development. Therefore, the DL model was deployed as a web application and validated with the unseen images (n = 114) and area under the receiver operating characteristic curve (AUC); other outcomes were calculated to assess the discrimination performance.

Results

The first baseline DL model had an AUC of 0.77 for PCNSL when evaluated with unseen images. The 2nd model with ridge regression regularization and the 3rd model with drop-out regularization increased an AUC of 0.83 and 0.84. In addition, the last model with data augmentation yielded an AUC of 0.57.

Conclusion

DL with regularization may provide useful diagnostic information to help doctors distinguish PCNSL from GBM.

Progress of radiological‑pathological workflows in the differential diagnosis between primary central nervous system lymphoma and high‑grade glioma (Review)

Primary central nervous system lymphoma (PCNSL) and high‑grade glioma (HGG) are distinct entities of the CNS with completely distinct treatments. The treatment of PCNSL is chemotherapy‑based, while surgery is the first choice for HGG. However, the clinical features of the two entities often overlap, and a clear pathological diagnosis is important for subsequent management, especially for the management of PCNSL. Stereotactic biopsy is recognized as one of the minimally invasive alternatives for evaluating the involvement of the CNS. However, in the case of limited tissue materials, the differential diagnosis between the two entities is still difficult. In addition, some patients are too ill to tolerate a needle biopsy. Therefore, combining imaging, histopathology and laboratory examinations is essential in order to make a clear diagnosis as soon as possible. The present study reviews the progress of comparative research on both imaging and laboratory tests based on the pathophysiological changes of the two entities, and proposes an integrative and optimized diagnostic process, with the purpose of building a better understanding for neurologists, hematologists, radiologists and pathologists.

Diagnostic Accuracy of the Diffusion-Weighted Imaging Method Used in Association With the Apparent Diffusion Coefficient for Differentiating Between Primary Central Nervous System Lymphoma and High-Grade Glioma: Systematic Review and Meta-Analysis

Background

It is difficult to differentiate between a few primary central nervous system lymphoma (PCNSL) and high-grade glioma (HGG) using conventional magnetic resonance imaging techniques. The purpose of this study is to explore whether diffusion-weighted imaging (DWI) can be effectively used to differentiate between these two types of tumors by analyzing the apparent diffusion coefficient (ADC).

Research Design and Methods

Data presented in Pubmed, Embase, Web of Science, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang Database, and China Science and Technology Journal Database (CQVIP) were analyzed. High-quality literature was included, and the quality was evaluated using the quality assessment of diagnostic accuracy studies-2 (QUADAS-2) tool, and the studies were based on the inclusion and exclusion rules. The pooled sensitivity, pooled specificity, pooled positive likelihood ratio (PLR), pooled negative likelihood ratio (NLR), pooled diagnostic odds ratio (DOR), area under the curve (AUC) of the summary operating characteristic curve (SROC), and corresponding 95% confidence interval (CI) were calculated using the bivariate mixed effect model. Meta-regression analysis and subgroup analysis were used to explore the sources of heterogeneity. The publication bias was evaluated by conducting Deek's test.

Results

In total, eighteen high-quality studies were included. The pooled sensitivity was 0.82 (95% CI: 0.75–0.88), the pooled specificity was 0.87 (95% CI: 0.84–0.90), the pooled positive likelihood ratio was 6.49 (95% CI: 5.06–8.32), the pooled NLR was 0.21 (95% CI: 0.14–0.30), the pooled DOR was 31.31 (95% CI: 18.55–52.86), and the pooled AUC was 0.90 (95% CI: 0.87–0.92). Sample size, language and country of publication, magnetic field strength, region of interest (ROI), and cut-off values of different types of ADC can potentially be the sources of heterogeneity. There was no publication bias in this meta-analysis.

Conclusions

The results obtained from the meta-analysis suggest that DWI is characterized by high diagnostic accuracy and thus can be effectively used for differentiating between PCNSL and HGG.

Radiomics-guided radiation therapy: opportunities and challenges

Radiomics is an advanced image-processing framework, which extracts image features and considers them as biomarkers towards personalized medicine. Applications include disease detection, diagnosis, prognosis, and therapy response assessment/prediction. As radiation therapy aims for further individualized treatments, radiomics could play a critical role in various steps before, during and after treatment. Elucidation of the concept of radiomics-guided radiation therapy (RGRT) is the aim of this review, attempting to highlight opportunities and challenges underlying the use of radiomics to guide clinicians and physicists towards more effective radiation treatments. This work identifies the value of RGRT in various steps of radiotherapy from patient selection to follow-up, and subsequently provides recommendations to improve future radiotherapy using quantitative imaging features.

Diagnostic efficacy of apparent diffusion coefficient measurements in differentiation of malignant intra-axial brain tumors

Background/aim

To evaluate diagnostic efficacy of the apparent diffusion coefficient measurements from tumor (ADCt) and tumor circumference hyperintensities (ADCtch) in different types of malignant intra-axial brain tumors.

Materials and methods

Between April 2013 and June 2017 , 125 patients (52 females (41.6%) and 73 males (58.4%); mean age: 53 years, age range: 14-81 years), who underwent diffusion-weighted imaging (DWI) with intracranial mass, were retrospectively evaluated. The mean ADC_t and ADC_tch values and ratios were measured.

Results

Of the 125 patients, 22 (17.6%) had a low-grade glioma (LGG), 55 (44%) had a high-grade glioma (HGG), 32 (25.6%) had metastasis, and 16 (12.8%) had lymphoma diagnosis. There was a statistically significant difference in LGG and HGG in terms of mean ADC_t and mean ADC_tch values, and ratios. ADC_tch values and ratios showed a statistically significant difference in the differentiation of HGG and metastasis and in the differentiation of HGG and lymphoma. According to ROC curve analysis, a cut-off value of 1.49 × 10−3 mm2/s for the mean ADC_tch value generated the best combination of 70% sensitivity and 71% specificity for differentiation of HGGs and metastasis. The mean ADC_tch value had the highest statistical predictive value for differentiation of HGGs and lymphoma with a sensitivity of 78% and a specificity of 76% for the optimal cut-off value of 0.82 × 10ˉ3 mm²/s.

Conclusion

The mean ADC_t ratio allowed reliable differentiation of LGG and high grade brain tumors, including HGGs, metastases, and lymphoma. The mean ADC_tch might be a better imaging biomarker in the differentiation of HHG from metastasis and lymphoma.