Background
Differentiation between glioblastoma and brain metastasis is highly important due to differing medical treatment strategies. While MRI is the modality of choice for the assessment of ...patients with brain tumors, differentiation between glioblastoma and solitary brain metastasis may be challenging due to their similar appearance on MRI.
Purpose
To differentiate between glioblastoma and brain metastasis subtypes using radiomics analysis based on conventional post‐contrast T1‐weighted (T1W) MRI.
Study Type
Retrospective.
Subjects
Data were acquired from 439 patients: 212 patients with glioblastoma and 227 patients with brain metastasis (breast, lung, and others).
Field Strength/Sequence
Post‐contrast 3D T1W gradient echo images, acquired with 1.5 and 3.0 T MR systems.
Assessment
Analysis included image preprocessing, segmentation of tumor area, and features extraction including: patients' clinical information, tumor location, first‐ and second‐order statistical, morphological, wavelet features, and bag‐of‐features. Following dimension reduction, classification was performed using various machine‐learning algorithms including support‐vector machine (SVM), k‐nearest neighbor, decision trees, and ensemble classifiers.
Statistical Tests
For classification, the data were divided into training (80%) and testing datasets (20%). Following optimization of the classifiers, mean sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated.
Results
For the testing dataset, the best results for differentiation of glioblastoma from brain metastasis were obtained using the SVM classifier with mean accuracy = 0.85, sensitivity = 0.86, specificity = 0.85, and AUC = 0.96. The best classification results between glioblastoma and brain metastasis subtypes were obtained using SVM classifier with mean accuracy = 0.85, 0.89, 0.75, 0.90; sensitivity = 1.00, 0.60, 0.57, 0.11; specificity = 0.76, 0.92, 0.87, 0.99; and AUC = 0.98, 0.81, 0.83, 0.57 for the glioblastoma, breast, lung, and other brain metastases, respectively.
Data Conclusion
Differentiation between glioblastoma and brain metastasis showed a high success rate based on postcontrast T1W MRI. Classification between glioblastoma and brain metastasis subtypes may require additional MR sequences with other tissue contrasts.
Level of Evidence: 1
Technical Efficacy: Stage 2
J. Magn. Reson. Imaging 2019;50:519–528.
Background
Deep‐learning is widely used for lesion classification. However, in the clinic patient data often has missing images.
Purpose
To evaluate the use of generated, duplicate and empty(black) ...images for replacing missing MRI data in AI brain tumor classification tasks.
Study Type
Retrospective.
Population
224 patients (local‐dataset; low‐grade‐glioma (LGG) = 37, high‐grade‐glioma (HGG) = 187) and 335 patients (public‐dataset (BraTS); LGG = 76, HGG = 259). The local‐dataset was divided into training (64), validation (16), and internal‐test‐data (20), while the public‐dataset was an independent test‐set.
Field Strength/Sequence
T1WI, T1WI+C, T2WI, and FLAIR images (1.5T/3.0T‐MR), obtained from different suppliers.
Assessment
Three image‐to‐image translation generative‐adversarial‐network (Pix2Pix‐GAN) models were trained on the local‐dataset, to generate T1WI, T2WI, and FLAIR images. The rating‐and‐preference‐judgment assessment was performed by three human‐readers (radiologist (MD) and two MRI‐technicians). Resnet152 was used for classification, and inference was performed on both datasets, with baseline input, and with missing data replaced by 1) generated images; 2) duplication of existing images; and 3) black images.
Statistical Tests
The similarity between the generated and the original images was evaluated using the peak‐signal‐to‐noise‐ratio (PSNR) and the structural‐similarity‐index‐measure (SSIM). Classification results were evaluated using accuracy, F1‐score and the Kolmogorov–Smirnov test and distance.
Results
For baseline‐state, the classification model reached to accuracy = 0.93,0.82 on the local and public‐datasets. For the missing‐data methods, high similarity was obtained between the generated and the original images with mean PSNR = 35.65,32.94 and SSIM = 0.87,0.91 on the local and public‐datasets; 39% of the generated‐images were labeled as real images by the human‐readers. The classification model using generated‐images to replace missing images produced the highest results with mean accuracy = 0.91,0.82 compared to 0.85,0.79 for duplicated and 0.77,0.68 for use of black images;
Data Conclusion
The feasibility for inference classification model on an MRI dataset with missing images using the Pix2pix‐GAN generated images, was shown. The stability and generalization ability of the model was demonstrated by producing consistent results on two independent datasets.
Level of Evidence
3
Technical Efficacy
Stage 5
Purpose: White-matter tract segmentation in patients with brain pathology can guide surgical planning and can be used for tissue integrity assessment. Recently, TractSeg was proposed for automatic ...tract segmentation in healthy subjects. The aim of this study was to assess the use of TractSeg for corticospinal-tract (CST) segmentation in a large cohort of patients with brain pathology and to evaluate its consistency in repeated measurements. Methods: A total of 649 diffusion-tensor-imaging scans were included, of them: 625 patients and 24 scans from 12 healthy controls (scanned twice for consistency assessment). Manual CST labeling was performed in all cases, and by 2 raters for the healthy subjects. Segmentation results were evaluated based on the Dice score. In order to evaluate consistency in repeated measurements, volume, Fractional Anisotropy (FA), and Mean Diffusivity (MD) values were extracted and correlated for the manual versus automatic methods. Results: For the automatic CST segmentation Dice scores of 0.63 and 0.64 for the training and testing datasets were obtained. Higher consistency between measurements was detected for the automatic segmentation, with between measurements correlations of volume = 0.92/0.65, MD = 0.94/0.75 for the automatic versus manual segmentation. Conclusions: The TractSeg method enables automatic CST segmentation in patients with brain pathology. Superior measurements consistency was detected for the automatic in comparison to manual fiber segmentation, which indicates an advantage when using this method for clinical and longitudinal studies.
Background
High‐grade gliomas (HGGs) induce both vasogenic edema and extensive infiltration of tumor cells, both of which present with similar appearance on conventional MRI. Using current ...radiological criteria, differentiation between these tumoral and nontumoral areas within the nonenhancing lesion area remains challenging.
Purpose
To use radiomics patch‐based analysis, based on conventional MRI, for the classification of the nonenhancing lesion area in patients with HGG into tumoral and nontumoral components.
Study Type
Prospective.
Subjects
In all, 179 MRI scans were obtained from 102 patients: 67 patients with HGG and 35 patients with brain metastases. A subgroup of 15 patients with HGG were scanned before and following administration of bevacizumab.
Field Strength/Sequence
Pre and postcontrast agent T1‐weighted‐imaging (WI), T2WI, FLAIR, diffusion‐tensor‐imaging (DTI), and dynamic‐contrast‐enhanced (DCE)‐MRI at 3T.
Assessment
A total of 225 histograms and gray‐level‐co‐occurrence matrix‐based features were extracted from the nonenhancing lesion area. Tumoral volumes of interest (VOIs) were defined at the peritumoral area in patients with HGG; nontumoral VOIs were defined in patients with brain metastasis. Twenty machine‐learning algorithms including support‐vector‐machine (SVM), k‐nearest neighbor, decision‐trees, and ensemble classifiers were tested. The best classifier was trained on the entire labeled data, and was used to classify the entire data.
Statistical Tests
Dimensional reduction was performed on the 225 features using principal component analysis. Classification results were evaluated based on the sensitivity, specificity, and accuracy of each of the 20 classifiers, first based on a training and testing dataset (80% of the labeled data) in a 5‐fold manner, and next by applying the best classifier to the validation data (the remaining 20% of the labeled data). Results were additionally evaluated by assessing differences in dynamic‐contrast‐enhanced plasma‐volume (vp) and volume‐transfer‐constant (ktrans) values between the two components using Mann–Whitney U‐test/t‐test.
Results
The best classification into tumoral and nontumoral lesion components was obtained using a linear SVM classifier, with average accuracy of 87%, sensitivity 86%, and specificity of 89% (for the training and testing data). Significantly higher vp and ktrans values (P < 0.0001) were detected in the tumoral compared to the nontumoral component. Preliminary classification results in a subgroup of patients treated with bevacizumab demonstrated a reduction mainly in the nontumoral component following administration of bevacizumab, enabling early assessment of disease progression in some patients.
Data Conclusion
A radiomics patch‐based analysis enables classification of the nonenhancing lesion area in patients with HGG. Preliminary results were promising and the proposed method has the potential to assist in clinical decision‐making and to improve therapy response assessment in patients with HGG.
Level of Evidence: 1
Technical Efficacy Stage 4
J. Magn. Reson. Imaging 2018;48:729–736.
Brain metastases are common in patients with advanced melanoma and constitute a major cause of morbidity and mortality. Between 40% and 60% of melanomas harbor BRAF mutations. Selective BRAF ...inhibitor therapy has yielded improvement in clinical outcome; however, genetic discordance between the primary lesion and the metastatic tumor has been shown to occur. Currently, the only way to characterize the genetic landscape of a brain metastasis is by tissue sampling, which carries risks and potential complications. The aim of this study was to investigate the use of radiomics analysis for non-invasive identification of BRAF mutation in patients with melanoma brain metastases, based on conventional magnetic resonance imaging (MRI) data. We applied a machine-learning method, based on MRI radiomics features for noninvasive characterization of the BRAF status of brain metastases from melanoma (BMM) and applied it to BMM patients from two tertiary neuro-oncological centers. All patients underwent surgical resection for BMM, and their BRAF mutation status was determined as part of their oncological work-up. Their routine preoperative MRI study was used for radiomics-based analysis in which 195 features were extracted and classified according to their BRAF status via a support vector machine. The BRAF status of 53 study patients, with 54 brain metastases (25 positive, 29 negative for BRAF mutation) was predicted with mean accuracy = 0.79 ± 0.13, mean precision = 0.77 ± 0.14, mean sensitivity = 0.72 ± 0.20, mean specificity = 0.83 ± 0.11 and with a 0.78 area under the receiver operating characteristic curve for positive BRAF mutation prediction. Radiomics-based noninvasive genetic characterization is feasible and should be further verified using large prospective cohorts.
Purpose
Non-small cell lung cancer (NSCLC) tends to metastasize to the brain. Between 10 and 60% of NSCLCs harbor an activating mutation in the epidermal growth-factor receptor (EGFR), which may be ...targeted with selective EGFR inhibitors. However, due to a high discordance rate between the molecular profile of the primary tumor and the brain metastases (BMs), identifying an individual patient’s EGFR status of the BMs necessitates tissue diagnosis via an invasive surgical procedure. We employed a deep learning (DL) method with the aim of noninvasive detection of the EGFR mutation status in NSCLC BM.
Methods
We retrospectively collected clinical, radiological, and pathological-molecular data of all the NSCLC patients who had been diagnosed with BMs and underwent resection of their BM during 2009–2019. The study population was then divided into two groups based upon EGFR mutational status. We further employed a DL technique to classify the two groups according to their preoperative magnetic resonance imaging features. Augmentation techniques, transfer learning approach, and post-processing of the predicted results were applied to overcome the relatively small cohort. Finally, we established the accuracy of our model in predicting EGFR mutation status of BM of NSCLC.
Results
Fifty-nine patients were included in the study, 16 patients harbored EGFR mutations. Our model predicted mutational status with mean accuracy of 89.8%, sensitivity of 68.7%, specificity of 97.7%, and a receiver operating characteristic curve value of 0.91 across the 5 validation datasets.
Conclusion
DL-based noninvasive molecular characterization is feasible, has high accuracy and should be further validated in large prospective cohorts.
T1-weighted MRI images are commonly used for volumetric assessment of brain structures. Magnetization prepared 2 rapid gradient echo (MP2RAGE) sequence offers superior gray (GM) and white matter (WM) ...contrast. This study aimed to quantitatively assess the agreement of whole brain tissue and deep GM (DGM) volumes obtained from MP2RAGE compared to the widely used MP-RAGE sequence. Twenty-nine healthy participants were included in this study. All subjects underwent a 3T MRI scan acquiring high-resolution 3D MP-RAGE and MP2RAGE images. Twelve participants were re-scanned after one year. The whole brain, as well as DGM segmentation, was performed using CAT12, volBrain, and FSL-FAST automatic segmentation tools based on the acquired images. Finally, contrast-to-noise ratio between WM and GM (CNR.sub.WG ), the agreement between the obtained tissue volumes, as well as scan-rescan variability of both sequences were explored. Significantly higher CNR.sub.WG was detected in MP2RAGE vs. MP-RAGE (Mean ± SD = 0.97 ± 0.04 vs. 0.8 ± 0.1 respectively; p<0.0001). Significantly higher total brain GM, and lower cerebrospinal fluid volumes were obtained from MP2RAGE vs. MP-RAGE based on all segmentation methods (p<0.05 in all cases). Whole-brain voxel-wise comparisons revealed higher GM tissue probability in the thalamus, putamen, caudate, lingual gyrus, and precentral gyrus based on MP2RAGE compared with MP-RAGE. Moreover, significantly higher WM probability was observed in the cerebellum, corpus callosum, and frontal-and-temporal regions in MP2RAGE vs. MP-RAGE. Finally, MP2RAGE showed a higher mean percentage of change in total brain GM compared to MP-RAGE. On the other hand, MP-RAGE demonstrated a higher overtime percentage of change in WM and DGM volumes compared to MP2RAGE. Due to its higher CNR, MP2RAGE resulted in reproducible brain tissue segmentation, and thus is a recommended method for volumetric imaging biomarkers for the monitoring of neurological diseases.
Differentiation between small-cell lung cancer (SCLC) from non-small-cell lung cancer (NSCLC) brain metastases is crucial due to the different clinical behaviors of the two tumor types. We propose ...the use of a deep learning and transfer learning approach based on conventional magnetic resonance imaging (MRI) for non-invasive classification of SCLC vs. NSCLC brain metastases. Sixty-nine patients with brain metastasis of lung cancer origin were included. Of them, 44 patients had NSCLC and 25 patients had SCLC. Classification was performed with EfficientNet architecture on crop images of lesion areas and based on post-contrast T1-weighted, T2-weighted and FLAIR imaging input data. Evaluation of the model was carried out in a 5-fold cross-validation manner, and based on accuracy, precision, recall, F1 score and area under the receiver operating characteristic curve. The best classification results were obtained with multiparametric MRI input data (T1WI+c+FLAIR+T2WI), with a mean overall accuracy of 0.90 ± 0.04, and F1 score of 0.92 ± 0.05 for NSCLC and 0.87 ± 0.08 for SCLC for the validation data and an accuracy of 0.87 ± 0.05, with an F1 score of 0.88 ± 0.05 for NSCLC and 0.85 ± 0.05 for SCLC for the test dataset. The proposed method provides an automatic noninvasive method for the classification of brain metastasis with high sensitivity and specificity for differentiation between NSCLC vs. SCLC brain metastases. It may be used as a diagnostic tool for improving decision-making in the treatment of patients with these metastases. Further studies on larger patient samples are required to validate the current results.
Purpose
Optic pathway gliomas (OPG) are low‐grade pilocytic astrocytomas accounting for 3‐5% of pediatric intracranial tumors. Accurate and quantitative follow‐up of OPG using magnetic resonance ...imaging (MRI) is crucial for therapeutic decision making, yet is challenging due to the complex shape and heterogeneous tissue pattern which characterizes these tumors. The aim of this study was to implement automatic methods for segmentation and classification of OPG and its components, based on MRI.
Methods
A total of 202 MRI scans from 29 patients with chiasmatic OPG scanned longitudinally were retrospectively collected and included in this study. Data included T2 and post‐contrast T1 weighted images. The entire tumor volume and its components were manually annotated by a senior neuro‐radiologist, and inter‐ and intra‐rater variability of the entire tumor volume was assessed in a subset of scans. Automatic tumor segmentation was performed using deep‐learning method with U‐Net+ResNet architecture. A fivefold cross‐validation scheme was used to evaluate the automatic results relative to manual segmentation. Voxel‐based classification of the tumor into enhanced, non‐enhanced, and cystic components was performed using fuzzy c‐means clustering.
Results
The results of the automatic tumor segmentation were: mean dice score = 0.736 ± 0.025, precision = 0.918 ± 0.014, and recall = 0.635 ± 0.039 for the validation data, and dice score = 0.761 ± 0.011, precision = 0.794 ± 0.028, and recall = 0.742 ± 0.012 for the test data. The accuracy of the voxel‐based classification of tumor components was 0.94, with precision = 0.89, 0.97, and 0.85, and recall = 1.00, 0.79, and 0.94 for the non‐enhanced, enhanced, and cystic components, respectively.
Conclusion
This study presents methods for automatic segmentation of chiasmatic OPG tumors and classification into the different components of the tumor, based on conventional MRI. Automatic quantitative longitudinal assessment of these tumors may improve radiological monitoring, facilitate early detection of disease progression and optimize therapy management.
Purpose
Low-grade gliomas (LGG) are classified into three distinct groups based on their IDH1 mutation and 1p/19q codeletion status, each of which is associated with a different clinical expression. ...The genomic sub-classification of LGG requires tumor sampling via neurosurgical procedures. The aim of this study was to evaluate the radiomics approach for noninvasive classification of patients with LGG and IDH mutation, based on their 1p/19q codeletion status, by testing different classifiers and assessing the contribution of the different MR contrasts.
Methods
Preoperative MRI scans of 47 patients diagnosed with LGG with IDH1-mutated tumors and a genetic analysis for 1p/19q deletion status were included in this study. A total of 152 features, including size, location and texture, were extracted from fluid-attenuated inversion recovery images,
T
2
-weighted images (WI) and post-contrast
T
1
WI
. Classification was performed using 17 machine learning classifiers. Results were evaluated by a fivefold cross-validation analysis.
Results
Radiomic analysis differentiated tumors with 1p/19q intact (
n
=
21
; astrocytomas) from those with 1p/19q codeleted (
n
=
26
; oligodendrogliomas). Best classification was obtained using the Ensemble Bagged Trees classifier, with sensitivity
=
92%, specificity
=
83% and accuracy
=
87%, and with area under the curve
=
0.87. Tumors with 1p/19q intact were larger than those with 1p/19q codeleted (
46.2
±
30.0
vs.
30.8
±
16.8
cc, respectively;
p
=
0.03
) and predominantly located to the left insula (
p
=
0.04
).
Conclusion
The proposed method yielded good discrimination between LGG with and without 1p/19q codeletion. Results from this study demonstrate the great potential of this method to aid decision-making in the clinical management of patients with LGG.