Reducing Gadolinium Contrast With Artificial Intelligence.

Gadolinium contrast is an important agent in magnetic resonance imaging (MRI), particularly in neuroimaging where it can help identify blood-brain barrier breakdown from an inflammatory, infectious, or neoplastic process. However, gadolinium contrast has several drawbacks, including nephrogenic systemic fibrosis, gadolinium deposition in the brain and bones, and allergic-like reactions. As computer hardware and technology continues to evolve, machine learning has become a possible solution for eliminating or reducing the dose of gadolinium contrast. This review summarizes the clinical uses of gadolinium contrast, the risks of gadolinium contrast, and state-of-the-art machine learning methods that have been applied to reduce or eliminate gadolinium contrast administration, as well as their current limitations, with a focus on neuroimaging applications. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 1.

Deep Learning to Predict Neonatal and Infant Brain Age from Myelination on Brain MRI Scans.

Background Assessment of appropriate brain myelination on T1- and T2-weighted MRI scans is based on gestationally corrected age (GCA) and requires subjective visual inspection of the brain with knowledge of normal myelination milestones. Purpose To develop a convolutional neural network (CNN) capable of estimating neonatal and infant GCA based on brain myelination on MRI scans. Materials and methods In this retrospective study from one academic medical center, brain MRI scans of patients aged 0-25 months with reported normal myelination were consecutively collected between January 1995 and June 2019. The GCA at MRI was manually calculated. After exclusion criteria were applied, T1- and T2-weighted MRI scans were preprocessed with skull stripping, linear registration, z scoring for normalization, and downsampling. A three-dimensional regression CNN was trained to predict GCA using mean absolute error (MAE) as its loss function. Attention maps were calculated using layer-wise relevance propagation. Models were validated on an external test set from the National Institutes of Health (NIH). Model MAEs were compared using Kruskal-Wallis and Mann-Whitney tests. Results A total of 518 neonates and infants (mean GCA, 67 weeks ± 33 [SD], 56% male) was included, comprising 469 T1-, 438 T2-, and 389 T1- and T2-weighted studies. Across 10 runs, MAEs of T1-, T2-, and T1- and T2-weighted networks were 9.8 ± 2.3, 9.1 ± 1.9, and 7.7 ± 1.7 weeks, respectively. Attention map analysis demonstrated increased network attention to the cerebellum, posterior white matter, and basal ganglia signal in neonates with GCA of less than 40 weeks and the anterior white matter signal in infants with GCA of more than 120 weeks, corresponding to the known progression of myelination. The T1- and T2-weighted network tested on the external NIH test set had an MAE of 9.1 weeks, which was reduced to 5.9 weeks with further training using half the external test set (P < .001). Conclusion A three-dimensional convolutional neural network can predict the gestationally corrected age of neonates and infants aged 0-25 months based on brain myelination patterns on T1- and T2-weighted MRI scans. © RSNA, 2022 Online supplemental material is available for this article.

Brain MRI Deep Learning and Bayesian Inference System Augments Radiology Resident Performance.

Automated quantitative and probabilistic medical image analysis has the potential to improve the accuracy and efficiency of the radiology workflow. We sought to determine whether AI systems for brain MRI diagnosis could be used as a clinical decision support tool to augment radiologist performance. We utilized previously developed AI systems that combine convolutional neural networks and expert-derived Bayesian networks to distinguish among 50 diagnostic entities on multimodal brain MRIs. We tested whether these systems could augment radiologist performance through an interactive clinical decision support tool known as Adaptive Radiology Interpretation and Education System (ARIES) in 194 test cases. Four radiology residents and three academic neuroradiologists viewed half of the cases unassisted and half with the results of the AI system displayed on ARIES. Diagnostic accuracy of radiologists for top diagnosis (TDx) and top three differential diagnosis (T3DDx) was compared with and without ARIES. Radiology resident performance was significantly better with ARIES for both TDx (55% vs 30%; P < .001) and T3DDx (79% vs 52%; P = 0.002), with the largest improvement for rare diseases (39% increase for T3DDx; P < 0.001). There was no significant difference between attending performance with and without ARIES for TDx (72% vs 69%; P = 0.48) or T3DDx (86% vs 89%; P = 0.39). These findings suggest that a hybrid deep learning and Bayesian inference clinical decision support system has the potential to augment diagnostic accuracy of non-specialists to approach the level of subspecialists for a large array of diseases on brain MRI.

Artificial Intelligence System Approaching Neuroradiologist-level Differential Diagnosis Accuracy at Brain MRI.

Background Although artificial intelligence (AI) shows promise across many aspects of radiology, the use of AI to create differential diagnoses for rare and common diseases at brain MRI has not been demonstrated. Purpose To evaluate an AI system for generation of differential diagnoses at brain MRI compared with radiologists. Materials and Methods This retrospective study tested performance of an AI system for probabilistic diagnosis in patients with 19 common and rare diagnoses at brain MRI acquired between January 2008 and January 2018. The AI system combines data-driven and domain-expertise methodologies, including deep learning and Bayesian networks. First, lesions were detected by using deep learning. Then, 18 quantitative imaging features were extracted by using atlas-based coregistration and segmentation. Third, these image features were combined with five clinical features by using Bayesian inference to develop probability-ranked differential diagnoses. Quantitative feature extraction algorithms and conditional probabilities were fine-tuned on a training set of 86 patients (mean age, 49 years ± 16 [standard deviation]; 53 women). Accuracy was compared with radiology residents, general radiologists, neuroradiology fellows, and academic neuroradiologists by using accuracy of top one, top two, and top three differential diagnoses in 92 independent test set patients (mean age, 47 years ± 18; 52 women). Results For accuracy of top three differential diagnoses, the AI system (91% correct) performed similarly to academic neuroradiologists (86% correct; P = .20), and better than radiology residents (56%; P < .001), general radiologists (57%; P < .001), and neuroradiology fellows (77%; P = .003). The performance of the AI system was not affected by disease prevalence (93% accuracy for common vs 85% for rare diseases; P = .26). Radiologists were more accurate at diagnosing common versus rare diagnoses (78% vs 47% across all radiologists; P < .001). Conclusion An artificial intelligence system for brain MRI approached overall top one, top two, and top three differential diagnoses accuracy of neuroradiologists and exceeded that of less-specialized radiologists. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Zaharchuk in this issue.

Emerging Applications of Artificial Intelligence in Neuro-Oncology.

Due to the exponential growth of computational algorithms, artificial intelligence (AI) methods are poised to improve the precision of diagnostic and therapeutic methods in medicine. The field of radiomics in neuro-oncology has been and will likely continue to be at the forefront of this revolution. A variety of AI methods applied to conventional and advanced neuro-oncology MRI data can already delineate infiltrating margins of diffuse gliomas, differentiate pseudoprogression from true progression, and predict recurrence and survival better than methods used in daily clinical practice. Radiogenomics will also advance our understanding of cancer biology, allowing noninvasive sampling of the molecular environment with high spatial resolution and providing a systems-level understanding of underlying heterogeneous cellular and molecular processes. By providing in vivo markers of spatial and molecular heterogeneity, these AI-based radiomic and radiogenomic tools have the potential to stratify patients into more precise initial diagnostic and therapeutic pathways and enable better dynamic treatment monitoring in this era of personalized medicine. Although substantial challenges remain, radiologic practice is set to change considerably as AI technology is further developed and validated for clinical use.

Esophageal Lichen Planus: Clinical and Radiographic Findings in Eight Patients.

OBJECTIVE: The purpose of this study is to present the clinical and radiographic findings of esophageal lichen planus. MATERIALS AND METHODS: A search of computerized medical records identified 15 patients with pathologic findings of esophageal lichen planus on endoscopic biopsy specimens. Three other patients had presumed esophageal lichen planus, although no biopsy specimens were obtained. Twelve of these 18 patients (67%) had double-contrast esophagography performed at our institution; for eight of the 12 patients (67%), the studies revealed abnormalities in the esophagus. These eight patients constituted our study group. The barium esophagrams and medical records of these eight patients were reviewed to determine the clinical, radiographic, and endoscopic findings of esophageal lichen planus as well as the treatment and patient outcome. RESULTS: All eight patients were women (median age, 66.5 years), and all eight presented with dysphagia (mean duration, 3.2 years). Four patients had previous lichen planus that involved the skin (n = 1), the oral cavity (n = 2), or both (n = 1), and one patient later had lichen planus that involved the vagina. Five patients had a small-caliber esophagus with diffuse esophageal narrowing. The remaining three patients had segmental strictures in the cervical (n = 1), upper thoracic (n = 1), and distal thoracic (n = 1) esophagus. CONCLUSION: Esophageal lichen planus typically occurs in older women with longstanding dysphagia and often develops in the absence of extraesophageal disease. Barium esophagrams may reveal a small-caliber esophagus or, less commonly, segmental esophageal strictures. Greater awareness of the radiographic findings of esophageal lichen planus hopefully will lead to earlier diagnosis and better management of this condition.

Position sensitivity in the visual word form area.

Seeing words involves the activity of neural circuitry within a small region in human ventral temporal cortex known as the visual word form area (VWFA). It is widely asserted that VWFA responses, which are essential for skilled reading, do not depend on the visual field position of the writing (position invariant). Such position invariance supports the hypothesis that the VWFA analyzes word forms at an abstract level, far removed from specific stimulus features. Using functional MRI pattern-classification techniques, we show that position information is encoded in the spatial pattern of VWFA responses. A right-hemisphere homolog (rVWFA) shows similarly position-sensitive responses. Furthermore, electrophysiological recordings in the human brain show position-sensitive VWFA response latencies. These findings show that position-sensitive information is present in the neural circuitry that conveys visual word form information to language areas. The presence of position sensitivity in the VWFA has implications for how word forms might be learned and stored within the reading circuitry.

Differential electrophysiological response during rest, self-referential, and non-self-referential tasks in human posteromedial cortex.

The electrophysiological basis for higher brain activity during rest and internally directed cognition within the human default mode network (DMN) remains largely unknown. Here we use intracranial recordings in the human posteromedial cortex (PMC), a core node within the DMN, during conditions of cued rest, autobiographical judgments, and arithmetic processing. We found a heterogeneous profile of PMC responses in functional, spatial, and temporal domains. Although the majority of PMC sites showed increased broad gamma band activity (30-180 Hz) during rest, some PMC sites, proximal to the retrosplenial cortex, responded selectively to autobiographical stimuli. However, no site responded to both conditions, even though they were located within the boundaries of the DMN identified with resting-state functional imaging and similarly deactivated during arithmetic processing. These findings, which provide electrophysiological evidence for heterogeneity within the core of the DMN, will have important implications for neuroimaging studies of the DMN.

Differences in educational opportunity predict white matter development.

Coarse measures of socioeconomic status, such as parental income or parental education, have been linked to differences in white matter development. However, these measures do not provide insight into specific aspects of an individual's environment and how they relate to brain development. On the other hand, educational intervention studies have shown that changes in an individual's educational context can drive measurable changes in their white matter. These studies, however, rarely consider socioeconomic factors in their results. In the present study, we examined the unique relationship between educational opportunity and white matter development, when controlling other known socioeconomic factors. To explore this question, we leveraged the rich demographic and neuroimaging data available in the ABCD study, as well the unique data-crosswalk between ABCD and the Stanford Education Data Archive (SEDA). We find that educational opportunity is related to accelerated white matter development, even when accounting for other socioeconomic factors, and that this relationship is most pronounced in white matter tracts associated with academic skills. These results suggest that the school a child attends has a measurable relationship with brain development for years to come.

Automated neonatal nnU-Net brain MRI extractor trained on a large multi-institutional dataset.

Brain extraction, or skull-stripping, is an essential data preprocessing step for machine learning approaches to brain MRI analysis. Currently, there are limited extraction algorithms for the neonatal brain. We aim to adapt an established deep learning algorithm for the automatic segmentation of neonatal brains from MRI, trained on a large multi-institutional dataset for improved generalizability across image acquisition parameters. Our model, ANUBEX (automated neonatal nnU-Net brain MRI extractor), was designed using nnU-Net and was trained on a subset of participants (N = 433) enrolled in the High-dose Erythropoietin for Asphyxia and Encephalopathy (HEAL) study. We compared the performance of our model to five publicly available models (BET, BSE, CABINET, iBEATv2, ROBEX) across conventional and machine learning methods, tested on two public datasets (NIH and dHCP). We found that our model had a significantly higher Dice score on the aggregate of both data sets and comparable or significantly higher Dice scores on the NIH (low-resolution) and dHCP (high-resolution) datasets independently. ANUBEX performs similarly when trained on sequence-agnostic or motion-degraded MRI, but slightly worse on preterm brains. In conclusion, we created an automatic deep learning-based neonatal brain extraction algorithm that demonstrates accurate performance with both high- and low-resolution MRIs with fast computation time.

The University of California San Francisco Adult Longitudinal Post-Treatment Diffuse Glioma MRI Dataset.

Supplemental material is available for this article.

The University of California San Francisco Brain Metastases Stereotactic Radiosurgery (UCSF-BMSR) MRI Dataset.

Supplemental material is available for this article.

White matter and literacy: A dynamic system in flux.

Cross-sectional studies have linked differences in white matter tissue properties to reading skills. However, past studies have reported a range of, sometimes conflicting, results. Some studies suggest that white matter properties act as individual-level traits predictive of reading skill, whereas others suggest that reading skill and white matter develop as a function of an individual's educational experience. In the present study, we tested two hypotheses: a) that diffusion properties of the white matter reflect stable brain characteristics that relate to stable individual differences in reading ability or b) that white matter is a dynamic system, linked with learning over time. To answer these questions, we examined the relationship between white matter and reading in a five-year longitudinal dataset and a series of large-scale, single-observation, cross-sectional datasets (N = 14,249 total participants). We find that gains in reading skill correspond to longitudinal changes in the white matter. However, in the cross-sectional datasets, we find no evidence for the hypothesis that individual differences in white matter predict reading skill. These findings highlight the link between dynamic processes in the white matter and learning.

Learning to see words.

Skilled reading requires recognizing written words rapidly; functional neuroimaging research has clarified how the written word initiates a series of responses in visual cortex. These responses are communicated to circuits in ventral occipitotemporal (VOT) cortex that learn to identify words rapidly. Structural neuroimaging has further clarified aspects of the white matter pathways that communicate reading signals between VOT and language systems. We review this circuitry, its development, and its deficiencies in poor readers. This review emphasizes data that measure the cortical responses and white matter pathways in individual subjects rather than group differences. Such methods have the potential to clarify why a child has difficulty learning to read and to offer guidance about the interventions that may be useful for that child.

3D U-Net for automated detection of multiple sclerosis lesions: utility of transfer learning from other pathologies.

Background and purpose: Deep learning algorithms for segmentation of multiple sclerosis (MS) plaques generally require training on large datasets. This manuscript evaluates the effect of transfer learning from segmentation of another pathology to facilitate use of smaller MS-specific training datasets. That is, a model trained for detection of one type of pathology was re-trained to identify MS lesions and active demyelination. Materials and methods: In this retrospective study using MRI exams from 149 patients spanning 4/18/2014 to 7/8/2021, 3D convolutional neural networks were trained with a variable number of manually-segmented MS studies. Models were trained for FLAIR lesion segmentation at a single timepoint, new FLAIR lesion segmentation comparing two timepoints, and enhancing (actively demyelinating) lesion segmentation on T1 post-contrast imaging. Models were trained either de-novo or fine-tuned with transfer learning applied to a pre-existing model initially trained on non-MS data. Performance was evaluated with lesionwise sensitivity and positive predictive value (PPV). Results: For single timepoint FLAIR lesion segmentation with 10 training studies, a fine-tuned model demonstrated improved performance [lesionwise sensitivity 0.55 ± 0.02 (mean ± standard error), PPV 0.66 ± 0.02] compared to a de-novo model (sensitivity 0.49 ± 0.02, p = 0.001; PPV 0.32 ± 0.02, p < 0.001). For new lesion segmentation with 30 training studies and their prior comparisons, a fine-tuned model demonstrated similar sensitivity (0.49 ± 0.05) and significantly improved PPV (0.60 ± 0.05) compared to a de-novo model (sensitivity 0.51 ± 0.04, p = 0.437; PPV 0.43 ± 0.04, p = 0.002). For enhancement segmentation with 20 training studies, a fine-tuned model demonstrated significantly improved overall performance (sensitivity 0.74 ± 0.06, PPV 0.69 ± 0.05) compared to a de-novo model (sensitivity 0.44 ± 0.09, p = 0.001; PPV 0.37 ± 0.05, p = 0.001). Conclusion: By fine-tuning models trained for other disease pathologies with MS-specific data, competitive models identifying existing MS plaques, new MS plaques, and active demyelination can be built with substantially smaller datasets than would otherwise be required to train new models.

Bayesian Networks in Radiology.

A Bayesian network is a graphical model that uses probability theory to represent relationships among its variables. The model is a directed acyclic graph whose nodes represent variables, such as the presence of a disease or an imaging finding. Connections between nodes express causal influences between variables as probability values. Bayesian networks can learn their structure (nodes and connections) and/or conditional probability values from data. Bayesian networks offer several advantages: (a) they can efficiently perform complex inferences, (b) reason from cause to effect or vice versa, (c) assess counterfactual data, (d) integrate observations with canonical ("textbook") knowledge, and (e) explain their reasoning. Bayesian networks have been employed in a wide variety of applications in radiology, including diagnosis and treatment planning. Unlike deep learning approaches, Bayesian networks have not been applied to computer vision. However, hybrid artificial intelligence systems have combined deep learning models with Bayesian networks, where the deep learning model identifies findings in medical images and the Bayesian network formulates and explains a diagnosis from those findings. One can apply a Bayesian network's probabilistic knowledge to integrate clinical and imaging findings to support diagnosis, treatment planning, and clinical decision-making. This article reviews the fundamental principles of Bayesian networks and summarizes their applications in radiology. Keywords: Bayesian Network, Machine Learning, Abdominal Imaging, Musculoskeletal Imaging, Breast Imaging, Neurologic Imaging, Radiology Education Supplemental material is available for this article. © RSNA, 2023.

High-resolution MR imaging of the orbit in patients with retinoblastoma.

Retinoblastoma is the most common intraocular childhood malignancy, with a prevalence of one in 18,000 children younger than 5 years old in the United States. In 80% of patients, retinoblastoma is diagnosed before the age of three, and in 95% of patients, retinoblastoma is diagnosed before the age of five. Although reports exist of retinoblastoma in adults, onset beyond 6 years of age is rare. Broadly, retinoblastoma may be classified into two groups: sporadic and heritable. In either case, the origin of the tumor is a biallelic mutation in primitive neuroepithelial cells. Although their details vary, several staging schemes are used to describe the extent of retinoblastoma according to the following four general criteria: intraocular location, extraocular (extraorbital) location, central nervous system disease, and systemic metastases. In the past decade, substantial changes have taken place in terms of staging and monitoring treatment in patients with retinoblastoma. Diagnosis and treatment of retinoblastoma involve a multidisciplinary approach, for which imaging is a vital component. Increasing awareness and concerns about the effects of radiation in patients with retinoblastoma have led to a shift away from external-beam radiation therapy and toward chemotherapy and locoregional treatment, as well as the establishment of magnetic resonance imaging as the most important imaging modality for diagnosis, staging, and treatment monitoring.

Combining radiomics and deep convolutional neural network features from preoperative MRI for predicting clinically relevant genetic biomarkers in glioblastoma.

Background: Glioblastoma is the most common primary brain malignancy, yet treatment options are limited, and prognosis remains guarded. Individualized tumor genetic assessment has become important for accurate prognosis and for guiding emerging targeted therapies. However, challenges remain for widespread tumor genetic testing due to costs and the need for tissue sampling. The aim of this study is to evaluate a novel artificial intelligence method for predicting clinically relevant genetic biomarkers from preoperative brain MRI in patients with glioblastoma. Methods: We retrospectively analyzed preoperative MRI data from 400 patients with glioblastoma, IDH-wildtype or WHO grade 4 astrocytoma, IDH mutant who underwent resection and genetic testing. Nine genetic biomarkers were assessed: hotspot mutations of IDH1 or TERT promoter, pathogenic mutations of TP53, PTEN, ATRX, or CDKN2A/B, MGMT promoter methylation, EGFR amplification, and combined aneuploidy of chromosomes 7 and 10. Models were developed to predict biomarker status from MRI data using radiomics features, convolutional neural network (CNN) features, and a combination of both. Results: Combined model performance was good for IDH1 and TERT promoter hotspot mutations, pathogenic mutations of ATRX and CDKN2A/B, and combined aneuploidy of chromosomes 7 and 10, with receiver operating characteristic area under the curve (ROC AUC) >0.85 and was fair for all other tested biomarkers with ROC AUC >0.7. Combined model performance was statistically superior to individual radiomics and CNN feature models for prediction chromosome 7 and 10 aneuploidy, MGMT promoter methylation, and PTEN mutation. Conclusions: Combining radiomics and CNN features from preoperative MRI yields improved noninvasive genetic biomarker prediction performance in patients with WHO grade 4 diffuse astrocytic gliomas.