Analyzing magnetic resonance imaging data from glioma patients using deep learning.
Ontology highlight
ABSTRACT:
SUBMITTER: Menze B
PROVIDER: S-EPMC8040671 | biostudies-literature | 2021 Mar
REPOSITORIES: biostudies-literature
Publications
Analyzing magnetic resonance imaging data from glioma patients using deep learning.
Menze Bjoern B Isensee Fabian F Wiest Roland R Wiestler Bene B Maier-Hein Klaus K Reyes Mauricio M Bakas Spyridon S
Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society 20201202
The quantitative analysis of images acquired in the diagnosis and treatment of patients with brain tumors has seen a significant rise in the clinical use of computational tools. The underlying technology to the vast majority of these tools are machine learning methods and, in particular, deep learning algorithms. This review offers clinical background information of key diagnostic biomarkers in the diagnosis of glioma, the most common primary brain tumor. It offers an overview of publicly availa ...[more]
Similar Datasets
Project description:PurposeTo explore the value of deep learning based on magnetic resonance imaging (MRI) in the classification of glioma subtypes.MethodsThis study retrospectively included 747 adult patients with surgically pathologically confirmed gliomas from a public database and 64 patients from our hospital. Patients were classified into IDH-wildtype (IDHwt) (490 cases), IDH-mutant/1p19q-noncodeleted (IDHmut-intact) (105 cases), and IDH-mutant/1p19q-codeleted (IDHmut-codel) (216 cases) based on their pathological findings, with the public database of patients were divided into training and validation sets, and patients from our hospital were used as an independent test set. The models were developed based on five categories of preoperative T1-weighted, T1-weighted gadolinium contrast-enhanced, T2-weighted and T2-weighted fluid-attenuated inversion recovery (T1w, T1c, T2w and FLAIR) magnetic resonance imaging (MRI) of four sequences and mixed imaging of the four sequences, respectively. The receiver operating characteristic curve (ROC), area under the curve (AUC) of the ROC were generated in the jupyter notebook tool using python language to evaluate the accuracy of the models in classification and comparing the predictive value of different MRI sequences.ResultsIDHwt, IDHmut-intact and IDHmut-codel were the best classified in the model containing only FLAIR sequences, with test set AUCs of 0.790, 0.737 and 0.820, respectively; and the worst classified in the model containing only T1w sequences, with test set AUCs of 0.621, 0.537 and 0.760, respectively.ConclusionWe have developed a set of models that can effectively classify glioma subtypes and that work best when only the FLAIR sequence model is included.
| S-EPMC11813795 | biostudies-literature
Project description:ObjectivesPhosphatase and tensin homolog (PTEN) mutation is an indicator of poor prognosis of low-grade and high-grade glioma. This study built a reliable model from multi-parametric magnetic resonance imaging (MRI) for predicting the PTEN mutation status in patients with glioma.MethodsIn this study, a total of 244 patients with glioma were retrospectively collected from our center (n = 77) and The Cancer Imaging Archive (n = 167). All patients were randomly divided into a training set (n = 170) and a validation set (n = 74). Three models were built from preoperative MRI for predicting PTEN status, including a radiomics model, a convolutional neural network (CNN) model, and an integrated model based on both radiomics and CNN features. The performance of each model was evaluated by accuracy and the area under the receiver operating characteristic curve (AUC).ResultsThe CNN model achieved an AUC of 0.84 and an accuracy of 0.81, which performed better than did the radiomics model, with an AUC of 0.83 and an accuracy of 0.66. Combining radiomics with CNN will further benefit the predictive performance (accuracy = 0.86, AUC = 0.91).ConclusionsThe combination of both the CNN and radiomics features achieved significantly higher performance in predicting the mutation status of PTEN in patients with glioma than did the radiomics or the CNN model alone.
| S-EPMC8521070 | biostudies-literature
Project description:BackgroundThe radiological differential diagnosis between tumor recurrence and radiation-induced necrosis (ie, pseudoprogression) is of paramount importance in the management of glioma patients.ObjectiveThis 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.MethodsIn 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.ResultsDNN 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 (P=.02 in sensitivity and P<.001 in specificity and accuracy).ConclusionsDeep 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.
| S-EPMC7708085 | biostudies-literature
Project description:BackgroundThe effect of diagnosing Graves' ophthalmopathy (GO) through traditional measurement and observation in medical imaging is not ideal. This study aimed to develop and validate deep learning (DL) models that could be applied to the diagnosis of GO based on magnetic resonance imaging (MRI) and compare them to traditional measurement and judgment of radiologists.MethodsA total of 199 clinically verified consecutive GO patients and 145 normal controls undergoing MRI were retrospectively recruited, of whom 240 were randomly assigned to the training group and 104 to the validation group. Areas of superior, inferior, medial, and lateral rectus muscles and all rectus muscles on coronal planes were calculated respectively. Logistic regression models based on areas of extraocular muscles were built to diagnose GO. The DL models named ResNet101 and Swin Transformer with T1-weighted MRI without contrast as input were used to diagnose GO and the results were compared to the radiologist's diagnosis only relying on MRI T1-weighted scans.ResultsAreas on the coronal plane of each muscle in the GO group were significantly greater than those in the normal group. In the validation group, the areas under the curve (AUCs) of logistic regression models by superior, inferior, medial, and lateral rectus muscles and all muscles were 0.897 [95% confidence interval (CI): 0.833-0.949], 0.705 (95% CI: 0.598-0.804), 0.799 (95% CI: 0.712-0.876), 0.681 (95% CI: 0.567-0.776), and 0.905 (95% CI: 0.843-0.955). ResNet101 and Swin Transformer achieved AUCs of 0.986 (95% CI: 0.977-0.994) and 0.936 (95% CI: 0.912-0.957), respectively. The accuracy, sensitivity, and specificity of ResNet101 were 0.933, 0.979, and 0.869, respectively. The accuracy, sensitivity, and specificity of Swin Transformer were 0.851, 0.817, and 0.898, respectively. The ResNet101 model yielded higher AUC than models of all muscles and radiologists (0.986 vs. 0.905, 0.818; P<0.001).ConclusionsThe DL models based on MRI T1-weighted scans could accurately diagnose GO, and the application of DL systems in MRI may improve radiologists' performance in diagnosing GO and early detection.
| S-EPMC11250345 | biostudies-literature
Project description:BackgroundMagnetic resonance spectroscopic imaging (MRSI) can be used in glioma patients to map the metabolic alterations associated with IDH1,2 mutations that are central criteria for glioma diagnosis. The aim of this study was to achieve super-resolution (SR) MRSI using deep learning to image tumor metabolism in patients with mutant IDH glioma.MethodsWe developed a deep learning method based on generative adversarial network (GAN) using Unet as generator network to upsample MRSI by a factor of 4. Neural networks were trained on simulated metabolic images from 75 glioma patients. The performance of deep neuronal networks was evaluated on MRSI data measured in 20 glioma patients and 10 healthy controls at 3T with a whole-brain 3D MRSI protocol optimized for detection of d-2-hydroxyglutarate (2HG). To further enhance structural details of metabolic maps we used prior information from high-resolution anatomical MR imaging. SR MRSI was compared to ground truth by Mann-Whitney U-test of peak signal-to-noise ratio (PSNR), structure similarity index measure (SSIM), feature-based similarity index measure (FSIM), and mean opinion score (MOS).ResultsDeep learning SR improved PSNR by 17%, SSIM by 5%, FSIM by 7%, and MOS by 30% compared to conventional interpolation methods. In mutant IDH glioma patients proposed method provided the highest resolution for 2HG maps to clearly delineate tumor margins and tumor heterogeneity.ConclusionsOur results indicate that proposed deep learning methods are effective in enhancing spatial resolution of metabolite maps. Patient results suggest that this may have great clinical potential for image guided precision oncology therapy.
| S-EPMC9332900 | biostudies-literature
Project description:BackgroundVentricular volumetry using a short-axis stack of two-dimensional (D) cine balanced steady-state free precession (bSSFP) sequences is crucial in any cardiac magnetic resonance imaging (MRI) examination. This task becomes particularly challenging in children due to multiple breath-holds.ObjectiveTo assess the diagnostic performance of accelerated 3-RR cine MRI sequences using deep learning reconstruction compared with standard 2-D cine bSSFP sequences.Material and methodsTwenty-nine consecutive patients (mean age 11 ± 5, median 12, range 1-17 years) undergoing cardiac MRI were scanned with a conventional segmented 2-D cine and a deep learning accelerated cine (three heartbeats) acquisition on a 1.5-tesla scanner. Short-axis volumetrics were performed (semi-)automatically in both datasets retrospectively by two experienced readers who visually assessed image quality employing a 4-point grading scale. Scan times and image quality were compared using the Wilcoxon rank-sum test. Volumetrics were assessed with linear regression and Bland-Altman analyses, and measurement agreement with intraclass correlation coefficient (ICC).ResultsMean acquisition time was significantly reduced with the 3-RR deep learning cine compared to the standard cine sequence (45.5 ± 13.8 s vs. 218.3 ± 44.8 s; P < 0.001). No significant differences in biventricular volumetrics were found. Left ventricular (LV) mass was increased in the deep learning cine compared with the standard cine sequence (71.4 ± 33.1 g vs. 69.9 ± 32.5 g; P < 0.05). All volumetric measurements had an excellent agreement with ICC > 0.9 except for ejection fraction (EF) (LVEF 0.81, RVEF 0.73). The image quality of deep learning cine images was decreased for end-diastolic and end-systolic contours, papillary muscles, and valve depiction (2.9 ± 0.5 vs. 3.5 ± 0.4; P < 0.05).ConclusionDeep learning cine volumetrics did not differ significantly from standard cine results except for LV mass, which was slightly overestimated with deep learning cine. Deep learning cine sequences result in a significant reduction in scan time with only slightly lower image quality.
| S-EPMC11377620 | biostudies-literature
Project description:ObjectivesIn adult diffuse glioma, preoperative detection of isocitrate dehydrogenase (IDH) status helps clinicians develop surgical strategies and evaluate patient prognosis. Here, we aim to identify an optimal machine-learning model for prediction of IDH genotyping by combining deep-learning (DL) signatures and conventional radiomics (CR) features as model predictors.MethodsIn this study, a total of 486 patients with adult diffuse gliomas were retrospectively collected from our medical center (n=268) and the public database (TCGA, n=218). All included patients were randomly divided into the training and validation sets by using nested 10-fold cross-validation. A total of 6,736 CR features were extracted from four MRI modalities in each patient, namely T1WI, T1CE, T2WI, and FLAIR. The LASSO algorithm was performed for CR feature selection. In each MRI modality, we applied a CNN+LSTM-based neural network to extract DL features and integrate these features into a DL signature after the fully connected layer with sigmoid activation. Eight classic machine-learning models were analyzed and compared in terms of their prediction performance and stability in IDH genotyping by combining the LASSO-selected CR features and integrated DL signatures as model predictors. In the validation sets, the prediction performance was evaluated by using accuracy and the area under the curve (AUC) of the receiver operating characteristics, while the model stability was analyzed by using the relative standard deviation of the AUC (RSDAUC). Subgroup analyses of DL signatures and CR features were also individually conducted to explore their independent prediction values.ResultsLogistic regression (LR) achieved favorable prediction performance (AUC: 0.920 ± 0.043, accuracy: 0.843 ± 0.044), whereas support vector machine with the linear kernel (l-SVM) displayed low prediction performance (AUC: 0.812 ± 0.052, accuracy: 0.821 ± 0.050). With regard to stability, LR also showed high robustness against data perturbation (RSDAUC: 4.7%). Subgroup analyses showed that DL signatures outperformed CR features (DL, AUC: 0.915 ± 0.054, accuracy: 0.835 ± 0.061, RSDAUC: 5.9%; CR, AUC: 0.830 ± 0.066, accuracy: 0.771 ± 0.051, RSDAUC: 8.0%), while DL and DL+CR achieved similar prediction results.ConclusionIn IDH genotyping, LR is a promising machine-learning classification model. Compared with CR features, DL signatures exhibit markedly superior prediction values and discriminative capability.
| S-EPMC10499353 | biostudies-literature
Project description:The goal of this study was to develop a deep learning-based algorithm to predict temporomandibular joint (TMJ) disc perforation based on the findings of magnetic resonance imaging (MRI) and to validate its performance through comparison with previously reported results. The study objects were obtained by reviewing medical records from January 2005 to June 2018. 299 joints from 289 patients were divided into perforated and non-perforated groups based on the existence of disc perforation confirmed during surgery. Experienced observers interpreted the TMJ MRI images to extract features. Data containing those features were applied to build and validate prediction models using random forest and multilayer perceptron (MLP) techniques, the latter using the Keras framework, a recent deep learning architecture. The area under the receiver operating characteristic (ROC) curve (AUC) was used to compare the performances of the models. MLP produced the best performance (AUC 0.940), followed by random forest (AUC 0.918) and disc shape alone (AUC 0.791). The MLP and random forest were also superior to previously reported results using MRI (AUC 0.808) and MRI-based nomogram (AUC 0.889). Implementing deep learning showed superior performance in predicting disc perforation in TMJ compared to conventional methods and previous reports.
| S-EPMC7988137 | biostudies-literature
Project description:Background and purposeSegmentation of oropharyngeal squamous cell carcinoma (OPSCC) is needed for radiotherapy planning. We aimed to segment the primary tumor for OPSCC on MRI using convolutional neural networks (CNNs). We investigated the effect of multiple MRI sequences as input and we proposed a semi-automatic approach for tumor segmentation that is expected to save time in the clinic.Materials and methodsWe included 171 OPSCC patients retrospectively from 2010 until 2015. For all patients the following MRI sequences were available: T1-weighted, T2-weighted and 3D T1-weighted after gadolinium injection. We trained a 3D UNet using the entire images and images with reduced context, considering only information within clipboxes around the tumor. We compared the performance using different combinations of MRI sequences as input. Finally, a semi-automatic approach by two human observers defining clipboxes around the tumor was tested. Segmentation performance was measured with Sørensen-Dice coefficient (Dice), 95th Hausdorff distance (HD) and Mean Surface Distance (MSD).ResultsThe 3D UNet trained with full context and all sequences as input yielded a median Dice of 0.55, HD of 8.7 mm and MSD of 2.7 mm. Combining all MRI sequences was better than using single sequences. The semi-automatic approach with all sequences as input yielded significantly better performance (p < 0.001): a median Dice of 0.74, HD of 4.6 mm and MSD of 1.2 mm.ConclusionReducing the amount of context around the tumor and combining multiple MRI sequences improved the segmentation performance. A semi-automatic approach was accurate and clinically feasible.
| S-EPMC8295848 | biostudies-literature
Project description:ObjectiveThe aim of the study is to develop and validate a deep learning model to predict the platinum sensitivity of patients with epithelial ovarian cancer (EOC) based on contrast-enhanced magnetic resonance imaging (MRI).MethodsIn this retrospective study, 93 patients with EOC who received platinum-based chemotherapy (≥4 cycles) and debulking surgery at the Sun Yat-sen Memorial Hospital from January 2011 to January 2020 were enrolled and randomly assigned to the training and validation cohorts (2:1). Two different models were built based on either the primary tumor or whole volume of the abdomen as the volume of interest (VOI) within the same cohorts, and then a pre-trained convolutional neural network Med3D (Resnet 10 version) was transferred to automatically extract 1,024 features from two MRI sequences (CE-T1WI and T2WI) of each patient to predict platinum sensitivity. The performance of the two models was compared.ResultsA total of 93 women (mean age, 50.5 years ± 10.5 [standard deviation]) were evaluated (62 in the training cohort and 31 in the validation cohort). The AUCs of the whole abdomen model were 0.97 and 0.98 for the training and validation cohorts, respectively, which was better than the primary tumor model (AUCs of 0.88 and 0.81 in the training and validation cohorts, respectively). In k-fold cross-validation and stratified analysis, the whole abdomen model maintained a stable performance, and the decision function value generated by the model was a prognostic indicator that successfully discriminates high- and low-risk recurrence patients.ConclusionThe non-manually segmented whole-abdomen deep learning model based on MRI exhibited satisfactory predictive performance for platinum sensitivity and may assist gynecologists in making optimal treatment decisions.
| S-EPMC9727155 | biostudies-literature