Impact of SUSAN Denoising and ComBat Harmonization on Machine Learning Model Performance for Malignant Brain Neoplasms.

BACKGROUND AND PURPOSE: Feature variability in radiomics studies due to technical and magnet strength parameters is well known and may be addressed through various pre-processing methods. However, very few studies have evaluated downstream impact of variable pre-processing on model classification performance in a multi-class setting. We sought to evaluate the impact of SUSAN denoising and ComBat harmonization on model classification performance. MATERIALS AND METHODS: A total of 493 cases (410 internal and 83 external dataset) of glioblastoma (GB), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL) underwent semi-automated 3D-segmentation post baseline image processing (BIP) consisting of resampling, realignment, co-registration, skull stripping and image normalization. Post BIP, two sets were generated, one with and another without SUSAN denoising (SD). Radiomics features were extracted from both datasets and batch corrected to produce four datasets: (a) BIP, (b) BIP with SD, (c) BIP with ComBat and (d) BIP with both SD and ComBat harmonization. Performance was then summarized for models using a combination of six feature selection techniques and six machine learning models across four mask-sequence combinations with features derived from one-three (multi-parametric) MRI sequences. RESULTS: Most top performing models on the external test set used BIP+SD derived features. Overall, use of SD and ComBat harmonization led to a slight but generally consistent improvement in model performance on the external test set. CONCLUSIONS: The use of image pre-processing steps such as SD and ComBat harmonization may be more useful in a multiinstitutional setting and improve model generalizability. Models derived from only T1-CE images showed comparable performance to models derived from multiparametric MRI.

High-Grade Astrocytoma with Piloid Features: A Dual Institutional Review of Imaging Findings of a Novel Entity.

High-grade astrocytoma with piloid features (HGAP) is a recently identified brain tumor characterized by a distinct DNA methylation profile. Predominantly located in the posterior fossa of adults, HGAP is notably prevalent in individuals with neurofibromatosis type 1. We present an image-centric review of HGAP and explore the association between HGAP and neurofibromatosis type 1. Data were collected from 8 HGAP patients treated at two tertiary care institutions between January 2020 and October 2023. Demographic details, clinical records, management, and tumor molecular profiles were analyzed. Tumor characteristics, including location and imaging features on MR imaging, were reviewed. Clinical or imaging features suggestive of neurofibromatosis 1 or the presence of NF1 gene alteration were documented. The mean age at presentation was 45.5 years (male/female = 5:3). Tumors were midline, localized in the posterior fossa (n = 4), diencephalic/thalamic (n = 2), and spinal cord (n = 2). HGAP lesions were T1 hypointense, T2-hyperintense, mostly without diffusion restriction, predominantly peripheral irregular enhancement with central necrosis (n = 3) followed by mixed heterogeneous enhancement (n = 2). Two NF1 mutation carriers showed signs of neurofibromatosis type 1 before HGAP diagnosis, with one diagnosed during HGAP evaluation, strengthening the HGAP-NF1 link, particularly in patients with posterior fossa masses. All tumors were IDH1 wild-type, often with ATRX, CDKN2A/B, and NF1 gene alteration. Six patients underwent surgical resection followed by adjuvant chemoradiation. Six patients were alive, and two died during the last follow-up. Histone H3 mutations were not detected in our cohort, such as the common H3K27M typically seen in diffuse midline gliomas, linked to aggressive clinical behavior and poor prognosis. HGAP lesions may involve the brain or spine and tend to be midline or paramedian in location. Underlying neurofibromatosis type 1 diagnosis or imaging findings are important diagnostic cues.

Integrating imaging and genomic data for the discovery of distinct glioblastoma subtypes: a joint learning approach.

Glioblastoma is a highly heterogeneous disease, with variations observed at both phenotypical and molecular levels. Personalized therapies would be facilitated by non-invasive in vivo approaches for characterizing this heterogeneity. In this study, we developed unsupervised joint machine learning between radiomic and genomic data, thereby identifying distinct glioblastoma subtypes. A retrospective cohort of 571 IDH-wildtype glioblastoma patients were included in the study, and pre-operative multi-parametric MRI scans and targeted next-generation sequencing (NGS) data were collected. L21-norm minimization was used to select a subset of 12 radiomic features from the MRI scans, and 13 key driver genes from the five main signal pathways most affected in glioblastoma were selected from the genomic data. Subtypes were identified using a joint learning approach called Anchor-based Partial Multi-modal Clustering on both radiomic and genomic modalities. Kaplan-Meier analysis identified three distinct glioblastoma subtypes: high-risk, medium-risk, and low-risk, based on overall survival outcome (p < 0.05, log-rank test; Hazard Ratio = 1.64, 95% CI 1.17-2.31, Cox proportional hazard model on high-risk and low-risk subtypes). The three subtypes displayed different phenotypical and molecular characteristics in terms of imaging histogram, co-occurrence of genes, and correlation between the two modalities. Our findings demonstrate the synergistic value of integrated radiomic signatures and molecular characteristics for glioblastoma subtyping. Joint learning on both modalities can aid in better understanding the molecular basis of phenotypical signatures of glioblastoma, and provide insights into the biological underpinnings of tumor formation and progression.

Repeated peripheral infusions of anti-EGFRvIII CAR T cells in combination with pembrolizumab show no efficacy in glioblastoma: a phase 1 trial.

We previously showed that chimeric antigen receptor (CAR) T-cell therapy targeting epidermal growth factor receptor variant III (EGFRvIII) produces upregulation of programmed death-ligand 1 (PD-L1) in the tumor microenvironment (TME). Here we conducted a phase 1 trial (NCT03726515) of CAR T-EGFRvIII cells administered concomitantly with the anti-PD1 (aPD1) monoclonal antibody pembrolizumab in patients with newly diagnosed, EGFRvIII+ glioblastoma (GBM) (n = 7). The primary outcome was safety, and no dose-limiting toxicity was observed. Secondary outcomes included median progression-free survival (5.2 months; 90% confidence interval (CI), 2.9-6.0 months) and median overall survival (11.8 months; 90% CI, 9.2-14.2 months). In exploratory analyses, comparison of the TME in tumors harvested before versus after CAR + aPD1 administration demonstrated substantial evolution of the infiltrating myeloid and T cells, with more exhausted, regulatory, and interferon (IFN)-stimulated T cells at relapse. Our study suggests that the combination of CAR T cells and PD-1 inhibition in GBM is safe and biologically active but, given the lack of efficacy, also indicates a need to consider alternative strategies.

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.

Non-Invasive Assessment of Isocitrate Dehydrogenase-Mutant Gliomas Using Optimized Proton Magnetic Resonance Spectroscopy on a Routine Clinical 3-Tesla MRI.

PURPOSE: The isocitrate dehydrogenase (IDH) mutation has become one of the most important prognostic biomarkers in glioma management, indicating better treatment response and prognosis. IDH mutations confer neomorphic activity leading to the conversion of alpha-ketoglutarate (α-KG) to 2-hydroxyglutarate (2HG). The purpose of this study was to investigate the clinical potential of proton MR spectroscopy (1H-MRS) in identifying IDH-mutant gliomas by detecting characteristic resonances of 2HG and its complex interplay with other clinically relevant metabolites. MATERIALS AND METHODS: Thirty-two patients with suspected infiltrative glioma underwent a single-voxel (SVS, n = 17) and/or single-slice-multivoxel (1H-MRSI, n = 15) proton MR spectroscopy (1H-MRS) sequence with an optimized echo-time (97 ms) on 3T-MRI. Spectroscopy data were analyzed using the linear combination (LC) model. Cramér-Rao lower bound (CRLB) values of <40% were considered acceptable for detecting 2HG and <20% for other metabolites. Immunohistochemical analyses for determining IDH mutational status were subsequently performed from resected tumor specimens and findings were compared with the results from spectral data. Mann-Whitney and chi-squared tests were performed to ascertain differences in metabolite levels between IDH-mutant and IDH-wild-type gliomas. Receiver operating characteristic (ROC) curve analyses were also performed. RESULTS: Data from eight cases were excluded due to poor spectral quality or non-tumor-related etiology, and final data analyses were performed from 24 cases. Of these cases, 9/12 (75%) were correctly identified as IDH-mutant or IDH-wildtype gliomas through SVS and 10/12 (83%) through 1H-MRSI with an overall concordance rate of 79% (19/24). The sensitivity, specificity, positive predictive value, and negative predictive value were 80%, 77%, 86%, and 70%, respectively. The metabolite 2HG was found to be significant in predicting IDH-mutant gliomas through the chi-squared test (p < 0.01). The IDH-mutant gliomas also had a significantly higher NAA/Cr ratio (1.20 ± 0.09 vs. 0.75 ± 0.12 p = 0.016) and lower Glx/Cr ratio (0.86 ± 0.078 vs. 1.88 ± 0.66; p = 0.029) than those with IDH wild-type gliomas. The areas under the ROC curves for NAA/Cr and Glx/Cr were 0.808 and 0.786, respectively. CONCLUSIONS: Noninvasive optimized 1H-MRS may be useful in predicting IDH mutational status and 2HG may serve as a valuable diagnostic and prognostic biomarker in patients with gliomas.

AI-based classification of three common malignant tumors in neuro-oncology: A multi-institutional comparison of machine learning and deep learning methods.

PURPOSE: To determine if machine learning (ML) or deep learning (DL) pipelines perform better in AI-based three-class classification of glioblastoma (GBM), intracranial metastatic disease (IMD) and primary CNS lymphoma (PCNSL). METHODOLOGY: Retrospective analysis included 502 cases for training (208 GBM, 67 PCNSL and 227 IMD), with external validation on 86 cases (27:27:32). Multiparametric MRI images (T1W, T2W, FLAIR, DWI and T1-CE) were co-registered, resampled, denoised and intensity normalized, followed by semiautomatic 3D segmentation of the enhancing tumor (ET) and peritumoral region (PTR). Model performance was assessed using several ML pipelines and 3D-convolutional neural networks (3D-CNN) using sequence specific masks, as well as combination of masks. All pipelines were trained and evaluated with 5-fold nested cross-validation on internal data followed by external validation using multi-class AUC. RESULTS: Two ML models achieved similar performance on test set, one using T2-ET and T2-PTR masks (AUC: 0.885, 95% CI: [0.816, 0.935] and another using T1-CE-ET and FLAIR-PTR mask (AUC: 0.878, CI: [0.804, 0.930]). The best performing DL models achieved an AUC of 0.854, (CI [0.774, 0.914]) on external data using T1-CE-ET and T2-PTR masks, followed by model derived from T1-CE-ET, ADC-ET and FLAIR-PTR masks (AUC: 0.851, CI [0.772, 0.909]). CONCLUSION: Both ML and DL derived pipelines achieved similar performance. T1-CE mask was used in three of the top four overall models. Additionally, all four models had some mask derived from PTR, either T2WI or FLAIR.

The Brain Tumor Segmentation (BraTS) Challenge 2023: Brain MR Image Synthesis for Tumor Segmentation (BraSyn).

Automated brain tumor segmentation methods have become well-established and reached performance levels offering clear clinical utility. These methods typically rely on four input magnetic resonance imaging (MRI) modalities: T1-weighted images with and without contrast enhancement, T2-weighted images, and FLAIR images. However, some sequences are often missing in clinical practice due to time constraints or image artifacts, such as patient motion. Consequently, the ability to substitute missing modalities and gain segmentation performance is highly desirable and necessary for the broader adoption of these algorithms in the clinical routine. In this work, we present the establishment of the Brain MR Image Synthesis Benchmark (BraSyn) in conjunction with the Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2023. The primary objective of this challenge is to evaluate image synthesis methods that can realistically generate missing MRI modalities when multiple available images are provided. The ultimate aim is to facilitate automated brain tumor segmentation pipelines. The image dataset used in the benchmark is diverse and multi-modal, created through collaboration with various hospitals and research institutions.

Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium.

PURPOSE: While the T2-FLAIR mismatch sign is highly specific for isocitrate dehydrogenase (IDH)-mutant, 1p/19q-noncodeleted astrocytomas among lower-grade gliomas, its utility in WHO grade 4 gliomas is not well-studied. We derived the partial T2-FLAIR mismatch sign as an imaging biomarker for IDH mutation in WHO grade 4 gliomas. METHODS: Preoperative MRI scans of adult WHO grade 4 glioma patients (n = 2165) from the multi-institutional ReSPOND (Radiomics Signatures for PrecisiON Diagnostics) consortium were analyzed. Diagnostic performance of the partial T2-FLAIR mismatch sign was evaluated. Subset analyses were performed to assess associations of imaging markers with overall survival (OS). RESULTS: One hundred twenty-one (5.6%) of 2165 grade 4 gliomas were IDH-mutant. Partial T2-FLAIR mismatch was present in 40 (1.8%) cases, 32 of which were IDH-mutant, yielding 26.4% sensitivity, 99.6% specificity, 80.0% positive predictive value, and 95.8% negative predictive value. Multivariate logistic regression demonstrated IDH mutation was significantly associated with partial T2-FLAIR mismatch (odds ratio [OR] 5.715, 95% CI [1.896, 17.221], p = 0.002), younger age (OR 0.911 [0.895, 0.927], p < 0.001), tumor centered in frontal lobe (OR 3.842, [2.361, 6.251], p < 0.001), absence of multicentricity (OR 0.173, [0.049, 0.612], p = 0.007), and presence of cystic (OR 6.596, [3.023, 14.391], p < 0.001) or non-enhancing solid components (OR 6.069, [3.371, 10.928], p < 0.001). Multivariate Cox analysis demonstrated cystic components (p = 0.024) and non-enhancing solid components (p = 0.003) were associated with longer OS, while older age (p < 0.001), frontal lobe center (p = 0.008), multifocality (p < 0.001), and multicentricity (p < 0.001) were associated with shorter OS. CONCLUSION: Partial T2-FLAIR mismatch sign is highly specific for IDH mutation in WHO grade 4 gliomas.

Skull Base Osteomyelitis: Historical Perspective, Diagnosis and Management Update.

SBO is a life-threatening disease that requires a high index of suspicion based on these patients complex underlying medical co-morbidities and clinician's acumen. Once a diagnosis is made, is it critical to communicate and work closely with other multidisciplinary teams (neuroradiology for appropriate choice of imaging study and interpretation; infectious disease for appropriate medical treatment and duration; internist to properly manage their underlying medical co-morbidities). Despite advances in imaging, the diagnosis is first made based on clinical judgment, appropriate culture, and tissue biopsy.

Assessment of treatment response to dendritic cell vaccine in patients with glioblastoma using a multiparametric MRI-based prediction model.

PURPOSE: Autologous tumor lysate-loaded dendritic cell vaccine (DCVax-L) is a promising treatment modality for glioblastomas. The purpose of this study was to investigate the potential utility of multiparametric MRI-based prediction model in evaluating treatment response in glioblastoma patients treated with DCVax-L. METHODS: Seventeen glioblastoma patients treated with standard-of-care therapy + DCVax-L were included. When tumor progression (TP) was suspected and repeat surgery was being contemplated, we sought to ascertain the number of cases correctly classified as TP + mixed response or pseudoprogression (PsP) from multiparametric MRI-based prediction model using histopathology/mRANO criteria as ground truth. Multiparametric MRI model consisted of predictive probabilities (PP) of tumor progression computed from diffusion and perfusion MRI-derived parameters. A comparison of overall survival (OS) was performed between patients treated with standard-of-care therapy + DCVax-L and standard-of-care therapy alone (external controls). Additionally, Kaplan-Meier analyses were performed to compare OS between two groups of patients using PsP, Ki-67, and MGMT promoter methylation status as stratification variables. RESULTS: Multiparametric MRI model correctly predicted TP + mixed response in 72.7% of cases (8/11) and PsP in 83.3% (5/6) with an overall concordance rate of 76.5% with final diagnosis as determined by histopathology/mRANO criteria. There was a significant concordant correlation coefficient between PP values and histopathology/mRANO criteria (r = 0.54; p = 0.026). DCVax-L-treated patients had significantly prolonged OS than those treated with standard-of-care therapy (22.38 ± 12.8 vs. 13.8 ± 9.5 months, p = 0.040). Additionally, glioblastomas with PsP, MGMT promoter methylation status, and Ki-67 values below median had longer OS than their counterparts. CONCLUSION: Multiparametric MRI-based prediction model can assess treatment response to DCVax-L in patients with glioblastoma.

Taming Glioblastoma in "Real Time": Integrating Multimodal Advanced Neuroimaging/AI Tools Towards Creating a Robust and Therapy Agnostic Model for Response Assessment in Neuro-Oncology.

The highly aggressive nature of glioblastoma carries a dismal prognosis despite aggressive multimodal therapy. Alternative treatment regimens, such as immunotherapies, are known to intensify the inflammatory response in the treatment field. Follow-up imaging in these scenarios often mimics disease progression on conventional MRI, making accurate evaluation extremely challenging. To this end, revised criteria for assessment of treatment response in high-grade gliomas were successfully proposed by the RANO Working Group to distinguish pseudoprogression from true progression, with intrinsic constraints related to the postcontrast T1-weighted MRI sequence. To address these existing limitations, our group proposes a more objective and quantifiable "treatment agnostic" model, integrating into the RANO criteria advanced multimodal neuroimaging techniques, such as diffusion tensor imaging (DTI), dynamic susceptibility contrast-perfusion weighted imaging (DSC-PWI), dynamic contrast enhanced (DCE)-MRI, MR spectroscopy, and amino acid-based positron emission tomography (PET) imaging tracers, along with artificial intelligence (AI) tools (radiomics, radiogenomics, and radiopathomics) and molecular information to address this complex issue of treatment-related changes versus tumor progression in "real-time", particularly in the early posttreatment window. Our perspective delineates the potential of incorporating multimodal neuroimaging techniques to improve consistency and automation for the assessment of early treatment response in neuro-oncology.

Validation of multiparametric MRI based prediction model in identification of pseudoprogression in glioblastomas.

BACKGROUND: Accurate differentiation of pseudoprogression (PsP) from tumor progression (TP) in glioblastomas (GBMs) is essential for appropriate clinical management and prognostication of these patients. In the present study, we sought to validate the findings of our previously developed multiparametric MRI model in a new cohort of GBM patients treated with standard therapy in identifying PsP cases. METHODS: Fifty-six GBM patients demonstrating enhancing lesions within 6 months after completion of concurrent chemo-radiotherapy (CCRT) underwent anatomical imaging, diffusion and perfusion MRI on a 3 T magnet. Subsequently, patients were classified as TP + mixed tumor (n = 37) and PsP (n = 19). When tumor specimens were available from repeat surgery, histopathologic findings were used to identify TP + mixed tumor (> 25% malignant features; n = 34) or PsP (< 25% malignant features; n = 16). In case of non-availability of tumor specimens, ≥ 2 consecutive conventional MRIs using mRANO criteria were used to determine TP + mixed tumor (n = 3) or PsP (n = 3). The multiparametric MRI-based prediction model consisted of predictive probabilities (PP) of tumor progression computed from diffusion and perfusion MRI derived parameters from contrast enhancing regions. In the next step, PP values were used to characterize each lesion as PsP or TP+ mixed tumor. The lesions were considered as PsP if the PP value was < 50% and TP+ mixed tumor if the PP value was ≥ 50%. Pearson test was used to determine the concordance correlation coefficient between PP values and histopathology/mRANO criteria. The area under ROC curve (AUC) was used as a quantitative measure for assessing the discriminatory accuracy of the prediction model in identifying PsP and TP+ mixed tumor. RESULTS: Multiparametric MRI model correctly predicted PsP in 95% (18/19) and TP+ mixed tumor in 57% of cases (21/37) with an overall concordance rate of 70% (39/56) with final diagnosis as determined by histopathology/mRANO criteria. There was a significant concordant correlation coefficient between PP values and histopathology/mRANO criteria (r = 0.56; p < 0.001). The ROC analyses revealed an accuracy of 75.7% in distinguishing PsP from TP+ mixed tumor. Leave-one-out cross-validation test revealed that 73.2% of cases were correctly classified as PsP and TP + mixed tumor. CONCLUSIONS: Our multiparametric MRI based prediction model may be helpful in identifying PsP in GBM patients.

Traumatic Brain Injury and Vision.

Traumatic brain injury disrupts the complex anatomy of the afferent and efferent visual pathways. Injury to the afferent pathway can result in vision loss, visual field deficits, and photophobia. Injury to the efferent pathway primarily causes eye movement abnormalities resulting in ocular misalignment and double vision. Injury to both the afferent and efferent systems can result in significant visual disability.

MRI-Based Radiomics Combined with Deep Learning for Distinguishing IDH-Mutant WHO Grade 4 Astrocytomas from IDH-Wild-Type Glioblastomas.

This study aimed to investigate the potential of quantitative radiomic data extracted from conventional MR images in discriminating IDH-mutant grade 4 astrocytomas from IDH-wild-type glioblastomas (GBMs). A cohort of 57 treatment-naïve patients with IDH-mutant grade 4 astrocytomas (n = 23) and IDH-wild-type GBMs (n = 34) underwent anatomical imaging on a 3T MR system with standard parameters. Post-contrast T1-weighted and T2-FLAIR images were co-registered. A semi-automatic segmentation approach was used to generate regions of interest (ROIs) from different tissue components of neoplasms. A total of 1050 radiomic features were extracted from each image. The data were split randomly into training and testing sets. A deep learning-based data augmentation method (CTGAN) was implemented to synthesize 200 datasets from the training sets. A total of 18 classifiers were used to distinguish two genotypes of grade 4 astrocytomas. From generated data using 80% training set, the best discriminatory power was obtained from core tumor regions overlaid on post-contrast T1 using the K-best feature selection algorithm and a Gaussian naïve Bayes classifier (AUC = 0.93, accuracy = 0.92, sensitivity = 1, specificity = 0.86, PR_AUC = 0.92). Similarly, high diagnostic performances were obtained from original and generated data using 50% and 30% training sets. Our findings suggest that conventional MR imaging-based radiomic features combined with machine/deep learning methods may be valuable in discriminating IDH-mutant grade 4 astrocytomas from IDH-wild-type GBMs.

Artificial Intelligence-Powered Clinical Decision Support and Simulation Platform for Radiology Trainee Education.

Technological tools can redesign traditional approaches to radiology education, for example, with simulation cases and via computer-generated feedback. In this study, we investigated the use of an AI-powered, Bayesian inference-based clinical decision support (CDS) software to provide automated "real-time" feedback to trainees during interpretation of clinical and simulation brain MRI examinations. Radiology trainees participated in sessions in which they interpreted 3 brain MRIs: two cases from a routine clinical worklist (one without and one with CDS) and a teaching file-based simulation case with CDS. The CDS software required trainees to input imaging features and differential diagnoses, after which inferred diagnoses were displayed, and the case was reviewed with an attending neuroradiologist. An observer timed each case, including time spent on education, and trainees completed a survey rating their confidence in their findings and the educational value of the case. Ten trainees reviewed 75 brain MRI examinations during 25 reading sessions. Trainees had slightly lower confidence in their findings and diagnosis and rated the educational value slightly higher for simulation cases with CDS compared to clinical cases without CDS (p < 0.05). There were no significant differences in ratings of clinical cases with or without CDS. No differences in overall timing were found among the reading scenarios. Simulation cases with "CDS-provided feedback" may improve the educational value of interpreting imaging studies at a workstation without adding additional time. Further investigation will help drive innovation in trainee education, which may be particularly relevant in this era of increasing remote work and asynchronous attending review.

Neuroimaging Patterns of Intracranial Infections: Meningitis, Cerebritis, and Their Complications.

Neuroimaging provides rapid, noninvasive visualization of central nervous system infections for optimal diagnosis and management. Generalizable and characteristic imaging patterns help radiologists distinguish different types of intracranial infections including meningitis and cerebritis from a variety of bacterial, viral, fungal, and/or parasitic causes. Here, we describe key radiologic patterns of meningeal enhancement and diffusion restriction through profiles of meningitis, cerebritis, abscess, and ventriculitis. We discuss various imaging modalities and recent diagnostic advances such as deep learning through a survey of intracranial pathogens and their radiographic findings. Moreover, we explore critical complications and differential diagnoses of intracranial infections.

Federated learning enables big data for rare cancer boundary detection.

Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.