An Attentive Multi-Modal CNN for Brain Tumor Radiogenomic Classification

MGMT promoter methylation status prediction using MRI scans? An extensive experimental evaluation of deep learning models

The number of studies on deep learning for medical diagnosis is expanding, and these systems are often claimed to outperform clinicians. However, only a few systems have shown medical efficacy. From this perspective, we examine a wide range of deep learning algorithms for the assessment of glioblastoma - a common brain tumor in older adults that is lethal. Surgery, chemotherapy, and radiation are the standard treatments for glioblastoma patients. The methylation status of the MGMT promoter, a specific genetic sequence found in the tumor, affects chemotherapy's effectiveness. MGMT promoter methylation improves chemotherapy response and survival in several cancers. MGMT promoter methylation is determined by a tumor tissue biopsy, which is then genetically tested. This lengthy and invasive procedure increases the risk of infection and other complications. Thus, researchers have used deep learning models to examine the tumor from brain MRI scans to determine the MGMT promoter's methylation state. We employ deep learning models and one of the largest public MRI datasets of 585 participants to predict the methylation status of the MGMT promoter in glioblastoma tumors using MRI scans. We test these models using Grad-CAM, occlusion sensitivity, feature visualizations, and training loss landscapes. Our results show no correlation between these two, indicating that external cohort data should be used to verify these models' performance to assure the accuracy and reliability of deep learning systems in cancer diagnosis.

Deep-learning and conventional radiomics to predict IDH genotyping status based on magnetic resonance imaging data in adult diffuse glioma

Objectives

In 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.

Methods

In 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.

Results

Logistic 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.

Conclusion

In IDH genotyping, LR is a promising machine-learning classification model. Compared with CR features, DL signatures exhibit markedly superior prediction values and discriminative capability.

Semi-supervised vision transformer with adaptive token sampling for breast cancer classification

Various imaging techniques combined with machine learning (ML) models have been used to build computer-aided diagnosis (CAD) systems for breast cancer (BC) detection and classification. The rise of deep learning models in recent years, represented by convolutional neural network (CNN) models, has pushed the accuracy of ML-based CAD systems to a new level that is comparable to human experts. Existing studies have explored the usage of a wide spectrum of CNN models for BC detection, and supervised learning has been the mainstream. In this study, we propose a semi-supervised learning framework based on the Vision Transformer (ViT). The ViT is a model that has been validated to outperform CNN models on numerous classification benchmarks but its application in BC detection has been rare. The proposed method offers a custom semi-supervised learning procedure that unifies both supervised and consistency training to enhance the robustness of the model. In addition, the method uses an adaptive token sampling technique that can strategically sample the most significant tokens from the input image, leading to an effective performance gain. We validate our method on two datasets with ultrasound and histopathology images. Results demonstrate that our method can consistently outperform the CNN baselines for both learning tasks. The code repository of the project is available at https://github.com/FeiYee/Breast-area-TWO.