Article Figures & Data
Figures
- Fig 1.
Visualization of computational image feature descriptors. A T1-weighted brain tumor section (A and B) is displayed, and feature visualizations (C–E) are given of LBP (C), HOG (D), and SIFT (E) descriptors. LBP quantifies local pixel structures through a binary coding scheme. HOG computes block-wise histogram gradients with multiple orientations. SIFT detects distributed key points with radius on tumor images. These multiparametric features create a rich image-driven data base to characterize tumors in MR imaging at different scales.
- Fig 2.
Linking subregional imaging to molecular profiles in glioblastoma. In this example, tumor subregions (B) are defined by jointly clustering on contrast-enhanced T1WI and T2WI (A). These subregions correspond to red (high T1WI and high T2WI), yellow (high T1WI and low T2WI), blue (low T1WI and high T2WI), and pink (low T1WI and low T2WI) areas. The defined tumor subregions enable quantitative spatial characterization, offering a means to noninvasively assess specific molecular activities (C) with enriched molecular pathways (D).
- Fig 3.
Illustration of the convolutional neural networks model using imaging and other biomedical data for brain tumor analysis. The convolutional neural networks model consists of multiple convolutional layers, pooling layers, and fully connected layers to learn an abstraction of the input data, such as imaging and clinical features for a variety of outcome evaluations.
Tables
Examples of quantitative features with their potential clinical relevance
Quantitative Feature Descriptors Potential Clinical Relevance Histogram of contrast-enhanced tumor MRI45 Distinguish molecular subtypes Contrast information between co-occurring subregions5 Survival predictor Pretreatment ADC histograms82 Indicator to bevacizumab treatment HOG34 Measure tumor microenvironment LBP27 Measure tumor microenvironment SIFT22 Measure tumor spatial characteristics
Jump to section
Related Articles
Cited By...
- From histology to macroscale function in the human amygdala
- Image-localized biopsy mapping of brain tumor heterogeneity: A single-center study protocol
- Evolving Role and Translation of Radiomics and Radiogenomics in Adult and Pediatric Neuro-Oncology
- Iodine Maps from Dual-Energy CT to Predict Extrathyroidal Extension and Recurrence in Papillary Thyroid Cancer Based on a Radiomics Approach
- Deep Learning for Pediatric Posterior Fossa Tumor Detection and Classification: A Multi-Institutional Study
- Deep Transfer Learning and Radiomics Feature Prediction of Survival of Patients with High-Grade Gliomas
- Artificial Intelligence in Obstetrics and Gynaecology: Is This the Way Forward?
- Radiogenomics in Medulloblastoma: Can the Human Brain Compete with Artificial Intelligence and Machine Learning?
- Integrating deep and radiomics features in cancer bioimaging
- Identifying Recurrent Malignant Glioma after Treatment Using Amide Proton Transfer-Weighted MR Imaging: A Validation Study with Image-Guided Stereotactic Biopsy
- MR Imaging-Based Radiomic Signatures of Distinct Molecular Subgroups of Medulloblastoma