Lower-grade gliomas (WHO grade II/III) have been classified into clinically relevant molecular subtypes based on and 1p/19q mutation status. The purpose was to investigate whether T2/FLAIR MRI features could distinguish between lower-grade glioma molecular subtypes. MRI scans from the TCGA/TCIA lower grade glioma database ( = 125) were evaluated by two independent neuroradiologists to assess (i) presence/absence of homogenous signal on T2WI; (ii) presence/absence of "T2-FLAIR mismatch" sign; (iii) sharp or indistinct lesion margins; and (iv) presence/absence of peritumoral edema. Metrics with moderate-substantial agreement underwent consensus review and were correlated with glioma molecular subtypes. Somatic mutation, DNA copy number, DNA methylation, gene expression, and protein array data from the TCGA lower-grade glioma database were analyzed for molecular-radiographic associations. A separate institutional cohort ( = 82) was analyzed to validate the T2-FLAIR mismatch sign. Among TCGA/TCIA cases, interreader agreement was calculated for lesion homogeneity [ = 0.234 (0.111-0.358)], T2-FLAIR mismatch sign [ = 0.728 (0.538-0.918)], lesion margins [ = 0.292 (0.135-0.449)], and peritumoral edema [ = 0.173 (0.096-0.250)]. All 15 cases that were positive for the T2-FLAIR mismatch sign were -mutant, 1p/19q non-codeleted tumors ( < 0.0001; PPV = 100%, NPV = 54%). Analysis of the validation cohort demonstrated substantial interreader agreement for the T2-FLAIR mismatch sign [ = 0.747 (0.536-0.958)]; all 10 cases positive for the T2-FLAIR mismatch sign were -mutant, 1p/19q non-codeleted tumors ( < 0.00001; PPV = 100%, NPV = 76%). Among lower-grade gliomas, T2-FLAIR mismatch sign represents a highly specific imaging biomarker for the -mutant, 1p/19q non-codeleted molecular subtype..

A novel application for non-thermal plasma is the induction of immunogenic cancer cell death for cancer immunotherapy. Cells undergoing immunogenic death emit danger signals which facilitate anti-tumor immune responses. Although pathways leading to immunogenic cell death are not fully understood; oxidative stress is considered to be part of the underlying mechanism. Here; we studied the interaction between dielectric barrier discharge plasma and cancer cells for oxidative stress-mediated immunogenic cell death. We assessed changes to the intracellular oxidative environment after plasma treatment and correlated it to emission of two danger signals: surface-exposed calreticulin and secreted adenosine triphosphate. Plasma-generated reactive oxygen and charged species were recognized as the major effectors of immunogenic cell death. Chemical attenuators of intracellular reactive oxygen species successfully abrogated oxidative stress following plasma treatment and modulated the emission of surface-exposed calreticulin. Secreted danger signals from cells undergoing immunogenic death enhanced the anti-tumor activity of macrophages. This study demonstrated that plasma triggers immunogenic cell death through oxidative stress pathways and highlights its potential development for cancer immunotherapy.

Accurate segmentation of different sub-regions of gliomas including peritumoral edema, necrotic core, enhancing and non-enhancing tumor core from multimodal MRI scans has important clinical relevance in diagnosis, prognosis and treatment of brain tumors. However, due to the highly heterogeneous appearance and shape, segmentation of the sub-regions is very challenging. Recent development using deep learning models has proved its effectiveness in the past several brain segmentation challenges as well as other semantic and medical image segmentation problems. Most models in brain tumor segmentation use a 2D/3D patch to predict the class label for the center voxel and variant patch sizes and scales are used to improve the model performance. However, it has low computation efficiency and also has limited receptive field. U-Net is a widely used network structure for end-to-end segmentation and can be used on the entire image or extracted patches to provide classification labels over the entire input voxels so that it is more efficient and expect to yield better performance with larger input size. Furthermore, instead of picking the best network structure, an ensemble of multiple models, trained on different dataset or different hyperparameters, can generally improve the segmentation performance. In this study we propose to use an ensemble of 3D U-Nets with different hyper-parameters for brain tumor segmentation. Preliminary results showed effectiveness of this model. In addition, we developed a linear model for survival prediction using extracted imaging and non-imaging features, which, despite the simplicity, can effectively reduce overfitting and regression errors.

The fraction of malignant histologic features in enhancing masses recurring after treatment for brain neoplasms can be predicted by using the rCBV fraction, with improved differentiation between recurrent neoplasm and TRN.

Imaging plays a pivotal role in the diagnostic process for many patients. With estimates of average diagnostic error rates ranging from 3% to 5%, there are approximately 40 million diagnostic errors involving imaging annually worldwide. The potential to improve diagnostic performance and reduce patient harm by identifying and learning from these errors is substantial. Yet these relatively high diagnostic error rates have persisted in our field despite decades of research and interventions. It may often seem as if diagnostic errors in radiology occur in a haphazard fashion. However, diagnostic problem solving in radiology is not a mysterious black box, and diagnostic errors are not random occurrences. Rather, diagnostic errors are predictable events with readily identifiable contributing factors, many of which are driven by how we think or related to the external environment. These contributing factors lead to both perceptual and interpretive errors. Identifying contributing factors is one of the keys to developing interventions that reduce or mitigate diagnostic errors. Developing a comprehensive process to identify diagnostic errors, analyze them to discover contributing factors and biases, and develop interventions based on the contributing factors is fundamental to learning from diagnostic error. Coupled with effective peer learning practices, supportive leadership, and a culture of quality, this process can unquestionably result in fewer diagnostic errors, improved patient outcomes, and increased satisfaction for all stakeholders. This article provides the foundational elements for implementing this type of process at a radiology practice, with examples to help radiologists and practice leaders achieve meaningful practice improvement. ©