Development and Validation of a Deep Learning-Based Model to Distinguish Glioblastoma from Solitary Brain Metastasis Using Conventional MR Images

Utilizing Radiomics of Peri-Lesional Edema in T2-FLAIR Subtraction Digital Images to Distinguish High-Grade Glial Tumors From Brain Metastasis

BACKGROUND

Differentiating high-grade glioma (HGG) and isolated brain metastasis (BM) is important for determining appropriate treatment. Radiomics, utilizing quantitative imaging features, offers the potential for improved diagnostic accuracy in this context.

PURPOSE

To differentiate high-grade (grade 4) glioma and BM using machine learning models from radiomics data obtained from T2-FLAIR digital subtraction images and the peritumoral edema area.

STUDY TYPE

Retrospective.

POPULATION

The study included 1287 patients. Of these, 602 were male and 685 were female. Of the 788 HGG patients included in the study, 702 had solitary masses. Of the 499 BM patients included in the study, 112 had solitary masses. Initially, the model was developed and tested on solitary masses. Subsequently, the model was developed and tested separately for all patients (solitary and multiple masses).

FIELD STRENGTH/SEQUENCE

Axial T2-weighted fast spin-echo sequence (T2WI) and T2-weighted fluid-attenuated inversion recovery sequence (T2-FLAIR), using 1.5-T and 3.0-T scanners.

ASSESSMENT

Radiomic features were extracted from digitally subtracted T2-FLAIR images in the area of peritumoral edema. The maximum relevance-minimum redundancy (mRMR) method was then used for dimensionality reduction. The naive Bayes algorithm was used in model development. The interpretability of the model was explored using SHapley Additive exPlanations (SHAP).

STATISTICAL TESTS

Chi-square test, one-way analysis of variance, and Kruskal-Wallis test were performed. The P values <0.05 were considered statistically significant. The performance metrics include area under curve (AUC), sensitivity (SENS), and specificity (SPEC).

RESULTS

The mean age of HGG patients was 61.4 ± 13.2 years and 61.7 ± 12.2 years for BM patients. In the external validation cohort, the model achieved AUC: 0.991, SENS: 0.983, and SPEC: 0.922. The external cohort results for patients with solitary lesions were AUC: 0.987, SENS: 0.950, and SPEC: 0.922.

DATA CONCLUSION

The artificial intelligence model, developed with radiomics data from the peritumoral edema area in T2-FLAIR digital subtraction images, might be able to differentiate isolated BM from HGG.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Multidimensional Proteomic Landscape Reveals Distinct Activated Pathways Between Human Brain Tumors

Brain metastases (BrMs) and gliomas are two typical human brain tumors with high incidence of mortalities and distinct clinical challenges, yet the understanding of these two types of tumors remains incomplete. Here, a multidimensional proteomic landscape of BrMs and gliomas to infer tumor-specific molecular pathophysiology at both tissue and plasma levels is presented. Tissue sample analysis reveals both shared and distinct characteristics of brain tumors, highlighting significant disparities between BrMs and gliomas with differentially activated upstream pathways of the PI3K-Akt signaling pathway that have been scarcely discussed previously. Novel proteins and phosphosites such as NSUN2, TM9SF3, and PRKCG_S330 are also detected, exhibiting a high correlation with reported clinical traits, which may serve as potential immunohistochemistry (IHC) biomarkers. Moreover, tumor-specific altered phosphosites and glycosites on FN1 are highlighted as potential therapeutic targets. Further validation of 110 potential noninvasive biomarkers yields three biomarker panels comprising a total of 19 biomarkers (including DES, VWF, and COL1A1) for accurate discrimination of two types of brain tumors and normal controls. In summary, this is a full-scale dataset of two typical human brain tumors, which serves as a valuable resource for advancing precision medicine in cancer patients through targeted therapy and immunotherapy.

Glioblastoma and Solitary Brain Metastasis: Differentiation by Integrating Demographic-MRI and Deep-Learning Radiomics Signatures

BACKGROUND

Studies have shown that deep-learning radiomics (DLR) could help differentiate glioblastoma (GBM) from solitary brain metastasis (SBM), but whether integrating demographic-MRI and DLR features can more accurately distinguish GBM from SBM remains uncertain.

PURPOSE

To construct and validate a demographic-MRI deep-learning radiomics nomogram (DDLRN) integrating demographic-MRI and DLR signatures to differentiate GBM from SBM preoperatively.

STUDY TYPE

Retrospective.

POPULATION

Two hundred and thirty-five patients with GBM (N = 115) or SBM (N = 120), randomly divided into a training cohort (90 GBM and 98 SBM) and a validation cohort (25 GBM and 22 SBM).

FIELD STRENGTH/SEQUENCE

Axial T2-weighted fast spin-echo sequence (T2WI), T2-weighted fluid-attenuated inversion recovery sequence (T2-FLAIR), and contrast-enhanced T1-weighted spin-echo sequence (CE-T1WI) using 1.5-T and 3.0-T scanners.

ASSESSMENT

The demographic-MRI signature was constructed with seven imaging features ("pool sign," "irregular ring sign," "regular ring sign," "intratumoral vessel sign," the ratio of the area of peritumoral edema to the enhanced tumor, the ratio of the lesion area on T2-FLAIR to CE-T1WI, and the tumor location) and demographic factors (age and sex). Based on multiparametric MRI, radiomics and deep-learning (DL) models, DLR signature, and DDLRN were developed and validated.

STATISTICAL TESTS

The Mann-Whitney U test, Pearson test, least absolute shrinkage and selection operator, and support vector machine algorithm were applied for feature selection and construction of radiomics and DL models.

RESULTS

DDLRN showed the best performance in differentiating GBM from SBM with area under the curves (AUCs) of 0.999 and 0.947 in the training and validation cohorts, respectively. Additionally, the DLR signature (AUC = 0.938) outperformed the radiomics and DL models, and the demographic-MRI signature (AUC = 0.775) was comparable to the T2-FLAIR radiomics and DL models in the validation cohort (AUC = 0.762 and 0.749, respectively).

DATA CONCLUSION

DDLRN integrating demographic-MRI and DLR signatures showed excellent performance in differentiating GBM from SBM.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 2.

Differentiation of glioblastoma from solitary brain metastasis using deep ensembles: Empirical estimation of uncertainty for clinical reliability

BACKGROUND AND OBJECTIVES

To develop a clinically reliable deep learning model to differentiate glioblastoma (GBM) from solitary brain metastasis (SBM) by providing predictive uncertainty estimates and interpretability.

METHODS

A total of 469 patients (300 GBM, 169 SBM) were enrolled in the institutional training set. Deep ensembles based on DenseNet121 were trained on multiparametric MRI. The model performance was validated in the external test set consisting of 143 patients (101 GBM, 42 SBM). Entropy values for each input were evaluated for uncertainty measurement; based on entropy values, the datasets were split to high- and low-uncertainty groups. In addition, entropy values of out-of-distribution (OOD) data from unknown class (257 patients with meningioma) were compared to assess uncertainty estimates of the model. The model interpretability was further evaluated by localization accuracy of the model.

RESULTS

On external test set, the area under the curve (AUC), accuracy, sensitivity and specificity of the deep ensembles were 0.83 (95 % confidence interval [CI] 0.76-0.90), 76.2 %, 54.8 % and 85.2 %, respectively. The performance was higher in the low-uncertainty group than in the high-uncertainty group, with AUCs of 0.91 (95 % CI 0.83-0.98) and 0.58 (95 % CI 0.44-0.71), indicating that assessment of uncertainty with entropy values ascertained reliable prediction in the low-uncertainty group. Further, deep ensembles classified a high proportion (90.7 %) of predictions on OOD data to be uncertain, showing robustness in dataset shift. Interpretability evaluated by localization accuracy provided further reliability in the "low-uncertainty and high-localization accuracy" subgroup, with an AUC of 0.98 (95 % CI 0.95-1.00).

CONCLUSIONS

Empirical assessment of uncertainty and interpretability in deep ensembles provides evidence for the robustness of prediction, offering a clinically reliable model in differentiating GBM from SBM.

Accuracy of Intra-Axial Brain Tumor Characterization in the Emergency MRI Reports: A Retrospective Human Performance Benchmarking Pilot Study

Demand for emergency neuroimaging is increasing. Even magnetic resonance imaging (MRI) is often performed outside office hours, sometimes revealing more uncommon entities like brain tumors. The scientific literature studying artificial intelligence (AI) methods for classifying brain tumors on imaging is growing, but knowledge about the radiologist's performance on this task is surprisingly scarce. Our study aimed to tentatively fill this knowledge gap. We hypothesized that the radiologist could classify intra-axial brain tumors at the emergency department with clinically acceptable accuracy. We retrospectively examined emergency brain MRI reports from 2013 to 2021, the inclusion criteria being (1) emergency brain MRI, (2) no previously known intra-axial brain tumor, and (3) suspicion of an intra-axial brain tumor on emergency MRI report. The tumor type suggestion and the final clinical diagnosis were pooled into groups: (1) glial tumors, (2) metastasis, (3) lymphoma, and (4) other tumors. The final study sample included 150 patients, of which 108 had histopathological tumor type confirmation. Among the patients with histopathological tumor type confirmation, the accuracy of the MRI reports in classifying the tumor type was 0.86 for gliomas against other tumor types, 0.89 for metastases, and 0.99 for lymphomas. We found the result encouraging, given the prolific need for emergency imaging.

Preoperative prediction of rectal Cancer staging combining MRI deep transfer learning, radiomics features, and clinical factors: accurate differentiation from stage T2 to T3

BACKGROUND

This study evaluates the efficacy of integrating MRI deep transfer learning, radiomic signatures, and clinical variables to accurately preoperatively differentiate between stage T2 and T3 rectal cancer.

METHODS

We included 361 patients with pathologically confirmed stage T2 or T3 rectal cancer, divided into a training set (252 patients) and a test set (109 patients) at a 7:3 ratio. The study utilized features derived from deep transfer learning and radiomics, with Spearman rank correlation and the Least Absolute Shrinkage and Selection Operator (LASSO) regression techniques to reduce feature redundancy. Predictive models were developed using Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), and Support Vector Machine (SVM), selecting the best-performing model for a comprehensive predictive framework incorporating clinical data.

RESULTS

After removing redundant features, 24 key features were identified. In the training set, the area under the curve (AUC)values for LR, RF, DT, and SVM were 0.867, 0.834, 0.900, and 0.944, respectively; in the test set, they were 0.847, 0.803, 0.842, and 0.910, respectively. The combined model, using SVM and clinical variables, achieved AUCs of 0.946 in the trainingset and 0.920 in the test set.

CONCLUSION

The study confirms the utility of a combined model of MRI deep transfer learning, radiomic features, and clinical factors for preoperative classification of stage T2 vs. T3 rectal cancer, offering significant technological support for precise diagnosis and potential clinical application.

Deep learning models for rapid discrimination of high-grade gliomas from solitary brain metastases using multi-plane T1-weighted contrast-enhanced (T1CE) images

Background

High-grade gliomas (HGG) and solitary brain metastases (SBM) are two common types of brain tumors in middle-aged and elderly patients. HGG and SBM display a high degree of similarity on magnetic resonance imaging (MRI) images. Consequently, differential diagnosis using preoperative MRI remains challenging. This study developed deep learning models that used pre-operative T1-weighted contrast-enhanced (T1CE) MRI images to differentiate between HGG and SBM before surgery.

Methods

By comparing various convolutional neural network models using T1CE image data from The First Medical Center of the Chinese PLA General Hospital and The Second People's Hospital of Yibin (Data collection for this study spanned from January 2016 to December 2023), it was confirmed that the GoogLeNet model exhibited the highest discriminative performance. Additionally, we evaluated the individual impact of the tumoral core and peritumoral edema regions on the network's predictive performance. Finally, we adopted a slice-based voting method to assess the accuracy of the validation dataset and evaluated patient prediction performance on an additional test dataset.

Results

The GoogLeNet model, in a five-fold cross-validation using multi-plane T1CE slices (axial, coronal, and sagittal) from 180 patients, achieved an average patient accuracy of 92.78%, a sensitivity of 95.56%, and a specificity of 90.00%. Moreover, on an external test set of 29 patients, the model achieved an accuracy of 89.66%, a sensitivity of 90.91%, and a specificity of 83.33%, with an area under the curve of 0.939 [95% confidence interval (CI): 0.842-1.000].

Conclusions

GoogLeNet performed better than previous methods at differentiating HGG from SBM, even for core and peritumoral edema in both. HGG and SBM could be fast screened using this end-to-end approach, improving workflow for both tumor treatments.

A comprehensive survey on the use of deep learning techniques in glioblastoma

Glioblastoma, characterized as a grade 4 astrocytoma, stands out as the most aggressive brain tumor, often leading to dire outcomes. The challenge of treating glioblastoma is exacerbated by the convergence of genetic mutations and disruptions in gene expression, driven by alterations in epigenetic mechanisms. The integration of artificial intelligence, inclusive of machine learning algorithms, has emerged as an indispensable asset in medical analyses. AI is becoming a necessary tool in medicine and beyond. Current research on Glioblastoma predominantly revolves around non-omics data modalities, prominently including magnetic resonance imaging, computed tomography, and positron emission tomography. Nonetheless, the assimilation of omic data-encompassing gene expression through transcriptomics and epigenomics-offers pivotal insights into patients' conditions. These insights, reciprocally, hold significant value in refining diagnoses, guiding decision- making processes, and devising efficacious treatment strategies. This survey's core objective encompasses a comprehensive exploration of noteworthy applications of machine learning methodologies in the domain of glioblastoma, alongside closely associated research pursuits. The study accentuates the deployment of artificial intelligence techniques for both non-omics and omics data, encompassing a range of tasks. Furthermore, the survey underscores the intricate challenges posed by the inherent heterogeneity of Glioblastoma, delving into strategies aimed at addressing its multifaceted nature.

Differentiation Between Glioblastoma and Metastatic Disease on Conventional MRI Imaging Using 3D-Convolutional Neural Networks: Model Development and Validation

RATIONALE AND OBJECTIVES

Imaging-based differentiation between glioblastoma (GB) and brain metastases (BM) remains challenging. Our aim was to evaluate the performance of 3D-convolutional neural networks (CNN) to address this binary classification problem.

MATERIALS AND METHODS

T1-CE, T2WI, and FLAIR 3D-segmented masks of 307 patients (157 GB and 150 BM) were generated post resampling, co-registration normalization and semi-automated 3D-segmentation and used for internal model development. Subsequent external validation was performed on 59 cases (27 GB and 32 BM) from another institution. Four different mask-sequence combinations were evaluated using area under the curve (AUC), precision, recall and F1-scores. Diagnostic performance of a neuroradiologist and a general radiologist, both without and with the model output available, was also assessed.

RESULTS

3D-model using the T1-CE tumor mask (TM) showed the highest performance [AUC 0.93 (95% CI 0.858-0.995)] on the external test set, followed closely by the model using T1-CE TM and FLAIR mask of peri-tumoral region (PTR) [AUC of 0.91 (95% CI 0.834-0.986)]. Models using T2WI masks showed robust performance on the internal dataset but lower performance on the external set. Both neuroradiologist and general radiologist showed improved performance with model output provided [AUC increased from 0.89 to 0.968 (p = 0.06) and from 0.78 to 0.965 (p = 0.007) respectively], the latter being statistically significant.

CONCLUSION

3D-CNNs showed robust performance for differentiating GB from BMs, with T1-CE TM, either alone or combined with FLAIR-PTR masks. Availability of model output significantly improved the accuracy of the general radiologist.

Classification of high-grade glioblastoma and single brain metastases using a new SCAT-inception model trained with MRI images

Background and objectives

Glioblastoma (GBM) and brain metastasis (MET) are the two most common intracranial tumors. However, the different pathogenesis of the two tumors leads to completely different treatment options. In terms of magnetic resonance imaging (MRI), GBM and MET are extremely similar, which makes differentiation by imaging extremely challenging. Therefore, this study explores an improved deep learning algorithm to assist in the differentiation of GBM and MET.

Materials and methods

For this study, axial contrast-enhanced T1 weight (ceT1W) MRI images from 321 cases of high-grade gliomas and solitary brain metastasis were collected. Among these, 251 out of 270 cases were selected for the experimental dataset (127 glioblastomas and 124 metastases), 207 cases were chosen as the training dataset, and 44 cases as the testing dataset. We designed a new deep learning algorithm called SCAT-inception (Spatial Convolutional Attention inception) and used five-fold cross-validation to verify the results.

Results

By employing the newly designed SCAT-inception model to predict glioblastomas and brain metastasis, the prediction accuracy reached 92.3%, and the sensitivity and specificity reached 93.5 and 91.1%, respectively. On the external testing dataset, our model achieved an accuracy of 91.5%, which surpasses other model performances such as VGG, UNet, and GoogLeNet.

Conclusion

This study demonstrated that the SCAT-inception architecture could extract more subtle features from ceT1W images, provide state-of-the-art performance in the differentiation of GBM and MET, and surpass most existing approaches.

Aided Diagnosis Model Based on Deep Learning for Glioblastoma, Solitary Brain Metastases, and Primary Central Nervous System Lymphoma with Multi-Modal MRI

(1) Background: Diagnosis of glioblastoma (GBM), solitary brain metastases (SBM), and primary central nervous system lymphoma (PCNSL) plays a decisive role in the development of personalized treatment plans. Constructing a deep learning classification network to diagnose GBM, SBM, and PCNSL with multi-modal MRI is important and necessary. (2) Subjects: GBM, SBM, and PCNSL were confirmed by histopathology with the multi-modal MRI examination (study from 1225 subjects, average age 53 years, 671 males), 3.0 T T2 fluid-attenuated inversion recovery (T2-Flair), and Contrast-enhanced T1-weighted imaging (CE-T1WI). (3) Methods: This paper introduces MFFC-Net, a classification model based on the fusion of multi-modal MRIs, for the classification of GBM, SBM, and PCNSL. The network architecture consists of parallel encoders using DenseBlocks to extract features from different modalities of MRI images. Subsequently, an L1-norm feature fusion module is applied to enhance the interrelationships among tumor tissues. Then, a spatial-channel self-attention weighting operation is performed after the feature fusion. Finally, the classification results are obtained using the full convolutional layer (FC) and Soft-max. (4) Results: The ACC of MFFC-Net based on feature fusion was 0.920, better than the radiomics model (ACC of 0.829). There was no significant difference in the ACC compared to the expert radiologist (0.920 vs. 0.924, p = 0.774). (5) Conclusions: Our MFFC-Net model could distinguish GBM, SBM, and PCNSL preoperatively based on multi-modal MRI, with a higher performance than the radiomics model and was comparable to radiologists.

The application value of deep learning in the background of precision medicine in glioblastoma

Introduction: Glioblastoma is a highly malignant central nervous system tumor, World Health Organization Ⅳ, glioblastoma is the most common primary malignancy, due to its own specificity and complexity, different patients often benefit from the current conventional treatment regimen because of different molecular subtypes, in the context of precision medicine, the application of deep learning to identify the salient features of tumors on brain imaging, prognostic predictive assessment combined with clinical data to maximize the benefits of each patient from the treatment regimen is a non-invasive and feasible regimen. Methods: We conducted a comprehensive review of the existing literature on the role of deep learning in glioblastomas, covering molecular classification and diagnosis, prognosis assessment. Results: Data based on a variety of magnetic resonance imaging sequences, genetic information, and clinical combinations enable noninvasive predictive tumor diagnosis of glioblastoma and assess overall survival and treatment response accuracy. For images, standardized image acquisition and data extraction techniques can be effectively translated into learning models for clinical practice. However, it must be recognized that interventions in the treatment of glioblastoma using deep learning are still in their infancy, and the robustness of the model is challenged, as the current total number of glioblastoma samples is insufficient for large-scale experimental methods, which is directly related to the difficulty of application of the model. Conclusion: Compared to radiomics and shallow machine learning, deep learning can be a more robust, non-invasive, and effective approach, providing more valuable information as clinicians develop personalized medical protocols for glioblastoma patients.

Using machine learning-based radiomics to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes

OBJECTIVE

To establish a machine learning-based radiomics model to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes, thereby achieving accurate preoperative classification.

MATERIALS AND METHODS

A retrospective analysis was conducted on MRI T1WI-enhanced images of 105 patients with glioma and 172 patients with solitary brain metastasis from lung cancer, which were confirmed pathologically. The patients were divided into the training group and validation group in an 8:2 ratio for image segmentation, extraction, and filtering; multiple layer perceptron (MLP), support vector machine (SVM), random forest (RF), and logistic regression (LR) were used for modeling; fivefold cross-validation was used to train the model; the validation group was used to evaluate and assess the predictive performance of the model, ROC curve was used to calculate the accuracy, sensitivity, and specificity of the model, and the area under curve (AUC) was used to assess the predictive performance of the model.

RESULTS

The accuracy and AUC of the MLP differentiation model for high-grade glioma and solitary brain metastasis in the validation group was 0.992, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.968, respectively. The accuracy and AUC for the MLP and SVM differentiation model for high-grade glioma and small cell lung cancer brain metastasis in the validation group was 0.966, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.929, respectively. The accuracy and AUC for the MLP differentiation model for high-grade glioma and non-small cell lung cancer brain metastasis in the validation group was 0.982, 0.999, respectively, while the sensitivity and specificity were 0.958, 1.000, respectively.

CONCLUSION

The application of machine learning-based radiomics has a certain clinical value in differentiating glioma from solitary brain metastasis from lung cancer and its subtypes. In the HGG/SBM and HGG/NSCLC SBM validation groups, the MLP model had the best diagnostic performance, while in the HGG/SCLC SBM validation group, the MLP and SVM models had the best diagnostic performance.

Artificial intelligence in neuroimaging of brain tumors: reality or still promise?

PURPOSE OF REVIEW

To provide an updated overview of artificial intelligence (AI) applications in neuro-oncologic imaging and discuss current barriers to wider clinical adoption.

RECENT FINDINGS

A wide variety of AI applications in neuro-oncologic imaging have been developed and researched, spanning tasks from pretreatment brain tumor classification and segmentation, preoperative planning, radiogenomics, prognostication and survival prediction, posttreatment surveillance, and differentiating between pseudoprogression and true disease progression. While earlier studies were largely based on data from a single institution, more recent studies have demonstrated that the performance of these algorithms are also effective on external data from other institutions. Nevertheless, most of these algorithms have yet to see widespread clinical adoption, given the lack of prospective studies demonstrating their efficacy and the logistical difficulties involved in clinical implementation.

SUMMARY

While there has been significant progress in AI and neuro-oncologic imaging, clinical utility remains to be demonstrated. The next wave of progress in this area will be driven by prospective studies measuring outcomes relevant to clinical practice and go beyond retrospective studies which primarily aim to demonstrate high performance.

Deep Learning With an Attention Mechanism for Differentiating the Origin of Brain Metastasis Using MR images

BACKGROUND

Brain metastasis (BM) is a serious neurological complication of cancer of different origins. The value of deep learning (DL) to identify multiple types of primary origins remains unclear.

PURPOSE

To distinguish primary site of BM and identify the best DL models.

STUDY TYPE

Retrospective.

POPULATION

A total of 449 BM derived from 214 patients (49.5% for female, mean age 58 years) (100 from small cell lung cancer [SCLC], 125 from non-small cell lung cancer [NSCLC], 116 from breast cancer [BC], and 108 from gastrointestinal cancer [GIC]) were included.

FIELD STRENGTH/SEQUENCE

A 3-T, T1 turbo spin echo (T1-TSE), T2-TSE, T2FLAIR-TSE, DWI echo-planar imaging (DWI-EPI) and contrast-enhanced T1-TSE (CE T1-TSE).

ASSESSMENT

Lesions were divided into training (n = 285, 153 patients), testing (n = 122, 93 patients), and independent testing cohorts (n = 42, 34 patients). Three-dimensional residual network (3D-ResNet), named 3D ResNet6 and 3D ResNet 18, was proposed for identifying the four origins based on single MRI and combined MRI (T1WI + T2-FLAIR + DWI, CE-T1WI + DWI, CE-T1WI + T2WI + DWI). DL model was used to distinguish lung cancer from non-lung cancer; then SCLC vs. NSCLC for lung cancer classification and BC vs. GIC for non-lung cancer classification was performed. A subjective visual analysis was implemented and compared with DL models. Gradient-weighted class activation mapping (Grad-CAM) was used to visualize the model by heatmaps.

STATISTICAL TESTS

The area under the receiver operating characteristics curve (AUC) assess each classification performance.

RESULTS

3D ResNet18 with Grad-CAM and AIC showed better performance than 3DResNet6, 3DResNet18 and the radiologist for distinguishing lung cancer from non-lung cancer, SCLC from NSCLC, and BC from GIC. For single MRI sequence, T1WI, DWI, and CE-T1WI performed best for lung cancer vs. non-lung cancer, SCLC vs. NSCLC, and BC vs. GIC classifications. The AUC ranged from 0.675 to 0.876 and from 0.684 to 0.800 regarding the testing and independent testing datasets, respectively. For combined MRI sequences, the combination of CE-T1WI + T2WI + DWI performed better for BC vs. GIC (AUCs of 0.788 and 0.848 on testing and independent testing datasets, respectively), while the combined MRI approach (T1WI + T2-FLAIR + DWI, CE-T1WI + DWI) could not achieve higher AUCs for lung cancer vs. non-lung cancer, SCLC vs. NSCLC. Grad-CAM helped for model visualization by heatmaps that focused on tumor regions.

DATA CONCLUSION

DL models may help to distinguish the origins of BM based on MRI data.

EVIDENCE LEVEL

3 TECHNICAL EFFICACY: Stage 2.

Discrimination Between Glioblastoma and Solitary Brain Metastasis Using Conventional MRI and Diffusion-Weighted Imaging Based on a Deep Learning Algorithm

This study aims to develop and validate a deep learning (DL) model to differentiate glioblastoma from single brain metastasis (BM) using conventional MRI combined with diffusion-weighted imaging (DWI). Preoperative conventional MRI and DWI of 202 patients with solitary brain tumor (104 glioblastoma and 98 BM) were retrospectively obtained between February 2016 and September 2022. The data were divided into training and validation sets in a 7:3 ratio. An additional 32 patients (19 glioblastoma and 13 BM) from a different hospital were considered testing set. Single-MRI-sequence DL models were developed using the 3D residual network-18 architecture in tumoral (T model) and tumoral + peritumoral regions (T&P model). Furthermore, the combination model based on conventional MRI and DWI was developed. The area under the receiver operating characteristic curve (AUC) was used to assess the classification performance. The attention area of the model was visualized as a heatmap by gradient-weighted class activation mapping technique. For the single-MRI-sequence DL model, the T2WI sequence achieved the highest AUC in the validation set with either T models (0.889) or T&P models (0.934). In the combination models of the T&P model, the model of DWI combined with T2WI and contrast-enhanced T1WI showed increased AUC of 0.949 and 0.930 compared with that of single-MRI sequences in the validation set, respectively. And the highest AUC (0.956) was achieved by combined contrast-enhanced T1WI, T2WI, and DWI. In the heatmap, the central region of the tumoral was hotter and received more attention than other areas and was more important for differentiating glioblastoma from BM. A conventional MRI-based DL model could differentiate glioblastoma from solitary BM, and the combination models improved classification performance.

Advances in the application of neuroinflammatory molecular imaging in brain malignancies

The prevalence of brain cancer has been increasing in recent decades, posing significant healthcare challenges. The introduction of immunotherapies has brought forth notable diagnostic imaging challenges for brain tumors. The tumor microenvironment undergoes substantial changes in induced immunosuppression and immune responses following the development of primary brain tumor and brain metastasis, affecting the progression and metastasis of brain tumors. Consequently, effective and accurate neuroimaging techniques are necessary for clinical practice and monitoring. However, patients with brain tumors might experience radiation-induced necrosis or other neuroinflammation. Currently, positron emission tomography and various magnetic resonance imaging techniques play a crucial role in diagnosing and evaluating brain tumors. Nevertheless, differentiating between brain tumors and necrotic lesions or inflamed tissues remains a significant challenge in the clinical diagnosis of the advancements in immunotherapeutics and precision oncology have underscored the importance of clinically applicable imaging measures for diagnosing and monitoring neuroinflammation. This review summarizes recent advances in neuroimaging methods aimed at enhancing the specificity of brain tumor diagnosis and evaluating inflamed lesions.

Deep learning regressor model based on nigrosome MRI in Parkinson syndrome effectively predicts striatal dopamine transporter-SPECT uptake

PURPOSE

Nigrosome imaging using susceptibility-weighted imaging (SWI) and dopamine transporter imaging using ¹²³I-2β-carbomethoxy-3β-(4-iodophenyl)-N-(3-fluoropropyl)-nortropane (¹²³I-FP-CIT) single-photon emission computerized tomography (SPECT) can evaluate Parkinsonism. Nigral hyperintensity from nigrosome-1 and striatal dopamine transporter uptake are reduced in Parkinsonism; however, quantification is only possible with SPECT. Here, we aimed to develop a deep-learning-based regressor model that can predict striatal ¹²³I-FP-CIT uptake on nigrosome magnetic resonance imaging (MRI) as a biomarker for Parkinsonism.

METHODS

Between February 2017 and December 2018, participants who underwent 3 T brain MRI including SWI and ¹²³I-FP-CIT SPECT based on suspected Parkinsonism were included. Two neuroradiologists evaluated the nigral hyperintensity and annotated the centroids of nigrosome-1 structures. We used a convolutional neural network-based regression model to predict striatal specific binding ratios (SBRs) measured via SPECT using the cropped nigrosome images. The correlation between measured and predicted SBRs was evaluated.

RESULTS

We included 367 participants (203 women (55.3%); age, 69.0 ± 9.2 [range, 39-88] years). Random data from 293 participants (80%) were used for training. In the test set (74 participants [20%]), the measured and predicted ¹²³I-FP-CIT SBRs were significantly lower with the loss of nigral hyperintensity (2.31 ± 0.85 vs. 2.44 ± 0.90) than with intact nigral hyperintensity (4.16 ± 1.24 vs. 4.21 ± 1.35, P < 0.01). The sorted measured ¹²³I-FP-CIT SBRs and the corresponding predicted values were significantly and positively correlated (ρ_c = 0.7443; 95% confidence interval, 0.6216-0.8314; P < 0.01).

CONCLUSION

A deep learning-based regressor model effectively predicted striatal ¹²³I-FP-CIT SBRs based on nigrosome MRI with high correlation using manually-measured values, enabling nigrosome MRI as a biomarker for nigrostriatal dopaminergic degeneration in Parkinsonism.

Artificial Intelligence in CT and MR Imaging for Oncological Applications

Cancer care increasingly relies on imaging for patient management. The two most common cross-sectional imaging modalities in oncology are computed tomography (CT) and magnetic resonance imaging (MRI), which provide high-resolution anatomic and physiological imaging. Herewith is a summary of recent applications of rapidly advancing artificial intelligence (AI) in CT and MRI oncological imaging that addresses the benefits and challenges of the resultant opportunities with examples. Major challenges remain, such as how best to integrate AI developments into clinical radiology practice, the vigorous assessment of quantitative CT and MR imaging data accuracy, and reliability for clinical utility and research integrity in oncology. Such challenges necessitate an evaluation of the robustness of imaging biomarkers to be included in AI developments, a culture of data sharing, and the cooperation of knowledgeable academics with vendor scientists and companies operating in radiology and oncology fields. Herein, we will illustrate a few challenges and solutions of these efforts using novel methods for synthesizing different contrast modality images, auto-segmentation, and image reconstruction with examples from lung CT as well as abdome, pelvis, and head and neck MRI. The imaging community must embrace the need for quantitative CT and MRI metrics beyond lesion size measurement. AI methods for the extraction and longitudinal tracking of imaging metrics from registered lesions and understanding the tumor environment will be invaluable for interpreting disease status and treatment efficacy. This is an exciting time to work together to move the imaging field forward with narrow AI-specific tasks. New AI developments using CT and MRI datasets will be used to improve the personalized management of cancer patients.

Unassisted Clinicians Versus Deep Learning-Assisted Clinicians in Image-Based Cancer Diagnostics: Systematic Review With Meta-analysis

BACKGROUND

A number of publications have demonstrated that deep learning (DL) algorithms matched or outperformed clinicians in image-based cancer diagnostics, but these algorithms are frequently considered as opponents rather than partners. Despite the clinicians-in-the-loop DL approach having great potential, no study has systematically quantified the diagnostic accuracy of clinicians with and without the assistance of DL in image-based cancer identification.

OBJECTIVE

We systematically quantified the diagnostic accuracy of clinicians with and without the assistance of DL in image-based cancer identification.

METHODS

PubMed, Embase, IEEEXplore, and the Cochrane Library were searched for studies published between January 1, 2012, and December 7, 2021. Any type of study design was permitted that focused on comparing unassisted clinicians and DL-assisted clinicians in cancer identification using medical imaging. Studies using medical waveform-data graphics material and those investigating image segmentation rather than classification were excluded. Studies providing binary diagnostic accuracy data and contingency tables were included for further meta-analysis. Two subgroups were defined and analyzed, including cancer type and imaging modality.

RESULTS

In total, 9796 studies were identified, of which 48 were deemed eligible for systematic review. Twenty-five of these studies made comparisons between unassisted clinicians and DL-assisted clinicians and provided sufficient data for statistical synthesis. We found a pooled sensitivity of 83% (95% CI 80%-86%) for unassisted clinicians and 88% (95% CI 86%-90%) for DL-assisted clinicians. Pooled specificity was 86% (95% CI 83%-88%) for unassisted clinicians and 88% (95% CI 85%-90%) for DL-assisted clinicians. The pooled sensitivity and specificity values for DL-assisted clinicians were higher than for unassisted clinicians, at ratios of 1.07 (95% CI 1.05-1.09) and 1.03 (95% CI 1.02-1.05), respectively. Similar diagnostic performance by DL-assisted clinicians was also observed across the predefined subgroups.

CONCLUSIONS

The diagnostic performance of DL-assisted clinicians appears better than unassisted clinicians in image-based cancer identification. However, caution should be exercised, because the evidence provided in the reviewed studies does not cover all the minutiae involved in real-world clinical practice. Combining qualitative insights from clinical practice with data-science approaches may improve DL-assisted practice, although further research is required.

TRIAL REGISTRATION

PROSPERO CRD42021281372; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=281372.

Differentiation of Glioblastoma and Brain Metastases by MRI-Based Oxygen Metabolomic Radiomics and Deep Learning

Glioblastoma (GB) and brain metastasis (BM) are the most frequent types of brain tumors in adults. Their therapeutic management is quite different and a quick and reliable initial characterization has a significant impact on clinical outcomes. However, the differentiation of GB and BM remains a major challenge in today's clinical neurooncology due to their very similar appearance in conventional magnetic resonance imaging (MRI). Novel metabolic neuroimaging has proven useful for improving diagnostic performance but requires artificial intelligence for implementation in clinical routines. Here; we investigated whether the combination of radiomic features from MR-based oxygen metabolism ("oxygen metabolic radiomics") and deep convolutional neural networks (CNNs) can support reliably pre-therapeutic differentiation of GB and BM in a clinical setting. A self-developed one-dimensional CNN combined with radiomic features from the cerebral metabolic rate of oxygen (CMRO₂) was clearly superior to human reading in all parameters for classification performance. The radiomic features for tissue oxygen saturation (mitoPO₂; i.e., tissue hypoxia) also showed better diagnostic performance compared to the radiologists. Interestingly, both the mean and median values for quantitative CMRO₂ and mitoPO₂ values did not differ significantly between GB and BM. This demonstrates that the combination of radiomic features and DL algorithms is more efficient for class differentiation than the comparison of mean or median values. Oxygen metabolic radiomics and deep neural networks provide insights into brain tumor phenotype that may have important diagnostic implications and helpful in clinical routine diagnosis.

Radiomics can differentiate high-grade glioma from brain metastasis: a systematic review and meta-analysis

OBJECTIVE

(1) To evaluate the diagnostic performance of radiomics in differentiating high-grade glioma from brain metastasis and how to improve the model. (2) To assess the methodological quality of radiomics studies and explore ways of embracing the clinical application of radiomics.

METHODS

Studies using radiomics to differentiate high-grade glioma from brain metastasis published by 26 July 2021 were systematically reviewed. Methodological quality and risk of bias were assessed using the Radiomics Quality Score (RQS) system and Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool, respectively. Pooled sensitivity and specificity of the radiomics model were also calculated.

RESULTS

Seventeen studies combining 1,717 patients were included in the systematic review, of which 10 studies without data leakage suspicion were employed for the quantitative statistical analysis. The average RQS was 5.13 (14.25% of total), with substantial or almost perfect inter-rater agreements. The inclusion of clinical features in the radiomics model was only reported in one study, as was the case for publicly available algorithm code. The pooled sensitivity and specificity were 84% (95% CI, 80-88%) and 84% (95% CI, 81-87%), respectively. The performances of feature extraction from the volume of interest (VOI) or (semi) automatic segmentation in the radiomics models were superior to those of protocols employing region of interest (ROI) or manual segmentation.

CONCLUSION

Radiomics can accurately differentiate high-grade glioma from brain metastasis. The adoption of standardized workflow to avoid potential data leakage as well as the integration of clinical features and radiomics are advised to consider in future studies.

KEY POINTS

• The pooled sensitivity and specificity of radiomics for differentiating high-grade gliomas from brain metastasis were 84% and 84%, respectively. • Avoiding potential data leakage by adopting an intensive and standardized workflow is essential to improve the quality and generalizability of the radiomics model. • The application of radiomics in combination with clinical features in differentiating high-grade gliomas from brain metastasis needs further validation.

Construction of VGG16 Convolution Neural Network (VGG16_CNN) Classifier with NestNet-Based Segmentation Paradigm for Brain Metastasis Classification

Brain metastases (BMs) happen often in patients with metastatic cancer (MC), requiring initial and precise diagnosis of BMs, which remains important for medical care preparation and radiotherapy prognostication. Nevertheless, the susceptibility of automated BM (ABMS) diagnosis is unfairly great for minute BMs, and integrating into medical exercises to distinguish true metastases (MtS) from false positives remains difficult. For enhancing BM classification execution, MtS localization is performed through the NestNet framework. Subsequent to segmentation, classification is performed by employing the VGG16 convolution neural network. A novel loss function is computed by employing the weighted softmax function (WSF) for enhancing minute MtS diagnosis and for calibrating susceptibility and particularity. The aim of this study was to merge temporal prior data for ABMS detection. The proffered VGG16_CNN is capable of differentiating positive MtS among MtS candidates with high confidence, which typically needs distinct specialist analysis or additional investigation, remaining specifically apt for specialist reinforcement in actual medical practice. The proffered VGG16_CNN framework can be correlated with three advanced methodologies (moU-Net, DSNet, and U-Net) concerning diverse criteria. It was observed that the proffered VGG16_CNN attained 93.74% accuracy, 92% precision, 92.1% recall, and 67.08% F1-score.

Inherited genetics of adult diffuse glioma and polygenic risk scores-a review

Knowledge about inherited and acquired genetics of adult diffuse glioma has expanded significantly over the past decade. Genomewide association studies (GWAS) stratified by histologic subtype identified six germline variants that were associated specifically with glioblastoma (GBM) and 12 that were associated with lower grade glioma. A GWAS performed using the 2016 WHO criteria, stratifying patients by IDH mutation and 1p/19q codeletion (as well as TERT promoter mutation), discovered that many of the known variants are associated with specific WHO glioma subtypes. In addition, the GWAS stratified by molecular group identified two additional novel regions: variants in D2HGDH that were associated with tumors that had an IDH mutation and a variant near FAM20C that was associated with tumors that had both IDH mutation and 1p/19q codeletion. The results of these germline associations have been used to calculate polygenic risk scores, from which to estimate relative and absolute risk of overall glioma and risk of specific glioma subtypes. We will review the concept of polygenic risk models and their potential clinical utility, as well as discuss the published adult diffuse glioma polygenic risk models. To date, these prior genetic studies have been done on European populations. Using the published glioma polygenic risk model, we show that the genetic associations published to date do not generalize across genetic ancestries, demonstrating that genetic studies need to be done on more diverse populations.

Image-Based Differentiation of Intracranial Metastasis From Glioblastoma Using Automated Machine Learning

Purpose

The majority of solitary brain metastases appear similar to glioblastomas (GBMs) on magnetic resonance imaging (MRI). This study aimed to develop and validate an MRI-based model to differentiate intracranial metastases from GBMs using automated machine learning.

Materials and Methods

Radiomics features from 354 patients with brain metastases and 354 with GBMs were used to build prediction algorithms based on T2-weighted images, contrast-enhanced (CE) T1-weighted images, or both. The data of these subjects were subjected to a nested 10-fold split in the training and testing groups to build the best algorithms using the tree-based pipeline optimization tool (TPOT). The algorithms were independently validated using data from 124 institutional patients with solitary brain metastases and 103 patients with GBMs from the cancer genome atlas.

Results

Three groups of models were developed. The average areas under the receiver operating characteristic curve (AUCs) were 0.856 for CE T1-weighted images, 0.976 for T2-weighted images, and 0.988 for a combination in the testing groups, and the AUCs of the groups of models in the independent validation were 0.687, 0.831, and 0.867, respectively. A total of 149 radiomics features were considered as the most valuable features for the differential diagnosis of GBMs and metastases.

Conclusion

The models established by TPOT can distinguish glioblastoma from solitary brain metastases well, and its non-invasiveness, convenience, and robustness make it potentially useful for clinical applications.

External Validation of Deep Learning Algorithms for Radiologic Diagnosis: A Systematic Review

Purpose

To assess generalizability of published deep learning (DL) algorithms for radiologic diagnosis.

Materials and Methods

In this systematic review, the PubMed database was searched for peer-reviewed studies of DL algorithms for image-based radiologic diagnosis that included external validation, published from January 1, 2015, through April 1, 2021. Studies using nonimaging features or incorporating non-DL methods for feature extraction or classification were excluded. Two reviewers independently evaluated studies for inclusion, and any discrepancies were resolved by consensus. Internal and external performance measures and pertinent study characteristics were extracted, and relationships among these data were examined using nonparametric statistics.

Results

Eighty-three studies reporting 86 algorithms were included. The vast majority (70 of 86, 81%) reported at least some decrease in external performance compared with internal performance, with nearly half (42 of 86, 49%) reporting at least a modest decrease (≥0.05 on the unit scale) and nearly a quarter (21 of 86, 24%) reporting a substantial decrease (≥0.10 on the unit scale). No study characteristics were found to be associated with the difference between internal and external performance.

Conclusion

Among published external validation studies of DL algorithms for image-based radiologic diagnosis, the vast majority demonstrated diminished algorithm performance on the external dataset, with some reporting a substantial performance decrease.Keywords: Meta-Analysis, Computer Applications-Detection/Diagnosis, Neural Networks, Computer Applications-General (Informatics), Epidemiology, Technology Assessment, Diagnosis, Informatics Supplemental material is available for this article. © RSNA, 2022.

Classification of Gliomas and Germinomas of the Basal Ganglia by Transfer Learning

Background

Germ cell tumors (GCTs) are neoplasms derived from reproductive cells, mostly occurring in children and adolescents at 10 to 19 years of age. Intracranial GCTs are classified histologically into germinomas and non-germinomatous germ cell tumors. Germinomas of the basal ganglia are difficult to distinguish based on symptoms or routine MRI images from gliomas, even for experienced neurosurgeons or radiologists. Meanwhile, intracranial germinoma has a lower incidence rate than glioma in children and adults. Therefore, we established a model based on pre-trained ResNet18 with transfer learning to better identify germinomas of the basal ganglia.

Methods

This retrospective study enrolled 73 patients diagnosed with germinoma or glioma of the basal ganglia. Brain lesions were manually segmented based on both T1C and T2 FLAIR sequences. The T1C sequence was used to build the tumor classification model. A 2D convolutional architecture and transfer learning were implemented. ResNet18 from ImageNet was retrained on the MRI images of our cohort. Class activation mapping was applied for the model visualization.

Results

The model was trained using five-fold cross-validation, achieving a mean AUC of 0.88. By analyzing the class activation map, we found that the model's attention was focused on the peri-tumoral edema region of gliomas and tumor bulk for germinomas, indicating that differences in these regions may help discriminate these tumors.

Conclusions

This study showed that the T1C-based transfer learning model could accurately distinguish germinomas from gliomas of the basal ganglia preoperatively.

Machine Learning Applications for Differentiation of Glioma from Brain Metastasis-A Systematic Review

Simple Summary

We present a systematic review of published reports on machine learning (ML) applications for the differentiation of gliomas from brain metastases by summarizing study characteristics, strengths, and pitfalls. Based on these findings, we present recommendations for future research in this field.

Abstract

Glioma and brain metastasis can be difficult to distinguish on conventional magnetic resonance imaging (MRI) due to the similarity of imaging features in specific clinical circumstances. Multiple studies have investigated the use of machine learning (ML) models for non-invasive differentiation of glioma from brain metastasis. Many of the studies report promising classification results, however, to date, none have been implemented into clinical practice. After a screening of 12,470 studies, we included 29 eligible studies in our systematic review. From each study, we aggregated data on model design, development, and best classifiers, as well as quality of reporting according to the TRIPOD statement. In a subset of eligible studies, we conducted a meta-analysis of the reported AUC. It was found that data predominantly originated from single-center institutions (n = 25/29) and only two studies performed external validation. The median TRIPOD adherence was 0.48, indicating insufficient quality of reporting among surveyed studies. Our findings illustrate that despite promising classification results, reliable model assessment is limited by poor reporting of study design and lack of algorithm validation and generalizability. Therefore, adherence to quality guidelines and validation on outside datasets is critical for the clinical translation of ML for the differentiation of glioma and brain metastasis.

Radiomics for precision medicine in glioblastoma

INTRODUCTION

Being the most common primary brain tumor, glioblastoma presents as an extremely challenging malignancy to treat with dismal outcomes despite treatment. Varying molecular epidemiology of glioblastoma between patients and intra-tumoral heterogeneity explains the failure of current one-size-fits-all treatment modalities. Radiomics uses machine learning to identify salient features of the tumor on brain imaging and promises patient-specific management in glioblastoma patients.

METHODS

We performed a comprehensive review of the available literature on studies investigating the role of radiomics and radiogenomics models for the diagnosis, stratification, prognostication as well as treatment planning and monitoring of glioblastoma.

RESULTS

Classifiers based on a combination of various MRI sequences, genetic information and clinical data can predict non-invasive tumor diagnosis, overall survival and treatment response with reasonable accuracy. However, the use of radiomics for glioblastoma treatment remains in infancy as larger sample sizes, standardized image acquisition and data extraction techniques are needed to develop machine learning models that can be translated effectively into clinical practice.

CONCLUSION

Radiomics has the potential to transform the scope of glioblastoma management through personalized medicine.

Neuroinflammation and immunoregulation in glioblastoma and brain metastases: Recent developments in imaging approaches

Brain tumors and brain metastases induce changes in brain tissue remodeling that lead to immunosuppression and trigger an inflammatory response within the tumor microenvironment. These immune and inflammatory changes can influence invasion and metastasis. Other neuroinflammatory and necrotic lesions may occur in patients with brain cancer or brain metastases as sequelae from treatment with radiotherapy. Glioblastoma (GBM) is the most aggressive primary malignant brain cancer in adults. Imaging methods such as positron emission tomography (PET) and different magnetic resonance imaging (MRI) techniques are highly valuable for the diagnosis and therapeutic evaluation of GBM and other malignant brain tumors. However, differentiating between tumor tissue and inflamed brain tissue with imaging protocols remains a challenge. Here, we review recent advances in imaging methods that have helped to improve the specificity of primary tumor diagnosis versus evaluation of inflamed and necrotic brain lesions. We also comment on advances in differentiating metastasis from neuroinflammation processes. Recent advances include the radiosynthesis of ¹⁸ F-FIMP, an L-type amino acid transporter 1 (LAT1)-specific PET probe that allows clearer differentiation between tumor tissue and inflammation compared to previous probes, and the combination of different advanced imaging protocols with the inclusion of radiomics and machine learning algorithms.