Selective and deceptive citation in the construction of dueling consensuses.

The COVID-19 pandemic provides a unique opportunity to study science communication and, in particular, the transmission of consensus. In this study, we show how "science communicators," writ large to include both mainstream science journalists and practiced conspiracy theorists, transform scientific evidence into two dueling consensuses using the effectiveness of masks as a case study. We do this by compiling one of the largest, hand-coded citation datasets of cross-medium science communication, derived from 5 million Twitter posts of people discussing masks. We find that science communicators selectively uplift certain published works while denigrating others to create bodies of evidence that support and oppose masks, respectively. Anti-mask communicators in particular often use selective and deceptive quotation of scientific work and criticize opposing science more than pro-mask communicators. Our findings have implications for scientists, science communicators, and scientific publishers, whose systems of sharing (and correcting) knowledge are highly vulnerable to what we term adversarial science communication.

Bavituximab Decreases Immunosuppressive Myeloid-Derived Suppressor Cells in Newly Diagnosed Glioblastoma Patients.

PURPOSE: We evaluated the efficacy of bavituximab-a mAb with anti-angiogenic and immunomodulatory properties-in newly diagnosed patients with glioblastoma (GBM) who also received radiotherapy and temozolomide. Perfusion MRI and myeloid-related gene transcription and inflammatory infiltrates in pre-and post-treatment tumor specimens were studied to evaluate on-target effects (NCT03139916). PATIENTS AND METHODS: Thirty-three adults with IDH--wild-type GBM received 6 weeks of concurrent chemoradiotherapy, followed by 6 cycles of temozolomide (C1-C6). Bavituximab was given weekly, starting week 1 of chemoradiotherapy, for at least 18 weeks. The primary endpoint was proportion of patients alive at 12 months (OS-12). The null hypothesis would be rejected if OS-12 was ≥72%. Relative cerebral blood flow (rCBF) and vascular permeability (Ktrans) were calculated from perfusion MRIs. Peripheral blood mononuclear cells and tumor tissue were analyzed pre-treatment and at disease progression using RNA transcriptomics and multispectral immunofluorescence for myeloid-derived suppressor cells (MDSC) and macrophages. RESULTS: The study met its primary endpoint with an OS-12 of 73% (95% confidence interval, 59%-90%). Decreased pre-C1 rCBF (HR, 4.63; P = 0.029) and increased pre-C1 Ktrans were associated with improved overall survival (HR, 0.09; P = 0.005). Pre-treatment overexpression of myeloid-related genes in tumor tissue was associated with longer survival. Post-treatment tumor specimens contained fewer immunosuppressive MDSCs (P = 0.01). CONCLUSIONS: Bavituximab has activity in newly diagnosed GBM and resulted in on-target depletion of intratumoral immunosuppressive MDSCs. Elevated pre-treatment expression of myeloid-related transcripts in GBM may predict response to bavituximab.

Followback Clusters, Satellite Audiences, and Bridge Nodes: Coengagement Networks for the 2020 US Election.

The 2020 United States (US) presidential election was - and has continued to be - the focus of pervasive and persistent mis- and disinformation spreading through our media ecosystems, including social media. This event has driven the collection and analysis of large, directed social network datasets, but such datasets can resist intuitive understanding. In such large datasets, the overwhelming number of nodes and edges present in typical representations create visual artifacts, such as densely overlapping edges and tightly-packed formations of low-degree nodes, which obscure many features of more practical interest. We apply a method, coengagement transformations, to convert such networks of social data into tractable images. Intuitively, this approach allows for parameterized network visualizations that make shared audiences of engaged viewers salient to viewers. Using the interpretative capabilities of this method, we perform an extensive case study of the 2020 United States presidential election on Twitter, contributing an empirical analysis of coengagement. By creating and contrasting different networks at different parameter sets, we define and characterize several structures in this discourse network, including bridging accounts, satellite audiences, and followback communities. We discuss the importance and implications of these empirical network features in this context. In addition, we release open-source code for creating coengagement networks from Twitter and other structured interaction data.

Combining interventions to reduce the spread of viral misinformation.

Misinformation online poses a range of threats, from subverting democratic processes to undermining public health measures. Proposed solutions range from encouraging more selective sharing by individuals to removing false content and accounts that create or promote it. Here we provide a framework to evaluate interventions aimed at reducing viral misinformation online both in isolation and when used in combination. We begin by deriving a generative model of viral misinformation spread, inspired by research on infectious disease. By applying this model to a large corpus (10.5 million tweets) of misinformation events that occurred during the 2020 US election, we reveal that commonly proposed interventions are unlikely to be effective in isolation. However, our framework demonstrates that a combined approach can achieve a substantial reduction in the prevalence of misinformation. Our results highlight a practical path forward as misinformation online continues to threaten vaccination efforts, equity and democratic processes around the globe.

Lung Nodule Malignancy Prediction in Sequential CT Scans: Summary of ISBI 2018 Challenge.

Lung cancer is by far the leading cause of cancer death in the US. Recent studies have demonstrated the effectiveness of screening using low dose CT (LDCT) in reducing lung cancer related mortality. While lung nodules are detected with a high rate of sensitivity, this exam has a low specificity rate and it is still difficult to separate benign and malignant lesions. The ISBI 2018 Lung Nodule Malignancy Prediction Challenge, developed by a team from the Quantitative Imaging Network of the National Cancer Institute, was focused on the prediction of lung nodule malignancy from two sequential LDCT screening exams using automated (non-manual) algorithms. We curated a cohort of 100 subjects who participated in the National Lung Screening Trial and had established pathological diagnoses. Data from 30 subjects were randomly selected for training and the remaining was used for testing. Participants were evaluated based on the area under the receiver operating characteristic curve (AUC) of nodule-wise malignancy scores generated by their algorithms on the test set. The challenge had 17 participants, with 11 teams submitting reports with method description, mandated by the challenge rules. Participants used quantitative methods, resulting in a reporting test AUC ranging from 0.698 to 0.913. The top five contestants used deep learning approaches, reporting an AUC between 0.87 - 0.91. The team's predictor did not achieve significant differences from each other nor from a volume change estimate (p =.05 with Bonferroni-Holm's correction).

Radiomics Repeatability Pitfalls in a Scan-Rescan MRI Study of Glioblastoma.

PURPOSE: To determine the influence of preprocessing on the repeatability and redundancy of radiomics features extracted using a popular open-source radiomics software package in a scan-rescan glioblastoma MRI study. MATERIALS AND METHODS: In this study, a secondary analysis of T2-weighted fluid-attenuated inversion recovery (FLAIR) and T1-weighted postcontrast images from 48 patients (mean age, 56 years [range, 22-77 years]) diagnosed with glioblastoma were included from two prospective studies (ClinicalTrials.gov NCT00662506 [2009-2011] and NCT00756106 [2008-2011]). All patients underwent two baseline scans 2-6 days apart using identical imaging protocols on 3-T MRI systems. No treatment occurred between scan and rescan, and tumors were essentially unchanged visually. Radiomic features were extracted by using PyRadiomics (https://pyradiomics.readthedocs.io/) under varying conditions, including normalization strategies and intensity quantization. Subsequently, intraclass correlation coefficients were determined between feature values of the scan and rescan. RESULTS: Shape features showed a higher repeatability than intensity (adjusted P < .001) and texture features (adjusted P < .001) for both T2-weighted FLAIR and T1-weighted postcontrast images. Normalization improved the overlap between the region of interest intensity histograms of scan and rescan (adjusted P < .001 for both T2-weighted FLAIR and T1-weighted postcontrast images), except in scans where brain extraction fails. As such, normalization significantly improves the repeatability of intensity features from T2-weighted FLAIR scans (adjusted P = .003 [z score normalization] and adjusted P = .002 [histogram matching]). The use of a relative intensity binning strategy as opposed to default absolute intensity binning reduces correlation between gray-level co-occurrence matrix features after normalization. CONCLUSION: Both normalization and intensity quantization have an effect on the level of repeatability and redundancy of features, emphasizing the importance of both accurate reporting of methodology in radiomics articles and understanding the limitations of choices made in pipeline design. Supplemental material is available for this article. © RSNA, 2020See also the commentary by Tiwari and Verma in this issue.

DeepNeuro: an open-source deep learning toolbox for neuroimaging.

Translating deep learning research from theory into clinical practice has unique challenges, specifically in the field of neuroimaging. In this paper, we present DeepNeuro, a Python-based deep learning framework that puts deep neural networks for neuroimaging into practical usage with a minimum of friction during implementation. We show how this framework can be used to design deep learning pipelines that can load and preprocess data, design and train various neural network architectures, and evaluate and visualize the results of trained networks on evaluation data. We present a way of reproducibly packaging data pre- and postprocessing functions common in the neuroimaging community, which facilitates consistent performance of networks across variable users, institutions, and scanners. We show how deep learning pipelines created with DeepNeuro can be concisely packaged into shareable Docker and Singularity containers with user-friendly command-line interfaces.

Multi-Institutional Assessment and Crowdsourcing Evaluation of Deep Learning for Automated Classification of Breast Density.

OBJECTIVE: We developed deep learning algorithms to automatically assess BI-RADS breast density. METHODS: Using a large multi-institution patient cohort of 108,230 digital screening mammograms from the Digital Mammographic Imaging Screening Trial, we investigated the effect of data, model, and training parameters on overall model performance and provided crowdsourcing evaluation from the attendees of the ACR 2019 Annual Meeting. RESULTS: Our best-performing algorithm achieved good agreement with radiologists who were qualified interpreters of mammograms, with a four-class κ of 0.667. When training was performed with randomly sampled images from the data set versus sampling equal number of images from each density category, the model predictions were biased away from the low-prevalence categories such as extremely dense breasts. The net result was an increase in sensitivity and a decrease in specificity for predicting dense breasts for equal class compared with random sampling. We also found that the performance of the model degrades when we evaluate on digital mammography data formats that differ from the one that we trained on, emphasizing the importance of multi-institutional training sets. Lastly, we showed that crowdsourced annotations, including those from attendees who routinely read mammograms, had higher agreement with our algorithm than with the original interpreting radiologists. CONCLUSION: We demonstrated the possible parameters that can influence the performance of the model and how crowdsourcing can be used for evaluation. This study was performed in tandem with the development of the ACR AI-LAB, a platform for democratizing artificial intelligence.

Evaluating the Use of rCBV as a Tumor Grade and Treatment Response Classifier Across NCI Quantitative Imaging Network Sites: Part II of the DSC-MRI Digital Reference Object (DRO) Challenge.

We have previously characterized the reproducibility of brain tumor relative cerebral blood volume (rCBV) using a dynamic susceptibility contrast magnetic resonance imaging digital reference object across 12 sites using a range of imaging protocols and software platforms. As expected, reproducibility was highest when imaging protocols and software were consistent, but decreased when they were variable. Our goal in this study was to determine the impact of rCBV reproducibility for tumor grade and treatment response classification. We found that varying imaging protocols and software platforms produced a range of optimal thresholds for both tumor grading and treatment response, but the performance of these thresholds was similar. These findings further underscore the importance of standardizing acquisition and analysis protocols across sites and software benchmarking.

Quantitative Imaging Informatics for Cancer Research.

PURPOSE: We summarize Quantitative Imaging Informatics for Cancer Research (QIICR; U24 CA180918), one of the first projects funded by the National Cancer Institute (NCI) Informatics Technology for Cancer Research program. METHODS: QIICR was motivated by the 3 use cases from the NCI Quantitative Imaging Network. 3D Slicer was selected as the platform for implementation of open-source quantitative imaging (QI) tools. Digital Imaging and Communications in Medicine (DICOM) was chosen for standardization of QI analysis outputs. Support of improved integration with community repositories focused on The Cancer Imaging Archive (TCIA). Priorities included improved capabilities of the standard, toolkits and tools, reference datasets, collaborations, and training and outreach. RESULTS: Fourteen new tools to support head and neck cancer, glioblastoma, and prostate cancer QI research were introduced and downloaded over 100,000 times. DICOM was amended, with over 40 correction proposals addressing QI needs. Reference implementations of the standard in a popular toolkit and standalone tools were introduced. Eight datasets exemplifying the application of the standard and tools were contributed. An open demonstration/connectathon was organized, attracting the participation of academic groups and commercial vendors. Integration of tools with TCIA was improved by implementing programmatic communication interface and by refining best practices for QI analysis results curation. CONCLUSION: Tools, capabilities of the DICOM standard, and datasets we introduced found adoption and utility within the cancer imaging community. A collaborative approach is critical to addressing challenges in imaging informatics at the national and international levels. Numerous challenges remain in establishing and maintaining the infrastructure of analysis tools and standardized datasets for the imaging community. Ideas and technology developed by the QIICR project are contributing to the NCI Imaging Data Commons currently being developed.

Bevacizumab Reduces Permeability and Concurrent Temozolomide Delivery in a Subset of Patients with Recurrent Glioblastoma.

PURPOSE: Targeting tumor blood vessels is an attractive therapy in glioblastoma (GBM), but the mechanism of action of these agents and how they modulate delivery of concomitant chemotherapy are not clear in humans. We sought to elucidate how bevacizumab modulates tumor vasculature and the impact those vascular changes have on drug delivery in patients with recurrent GBM. EXPERIMENTAL DESIGN: Temozolomide was labeled with [11C], and serial PET-MRI scans were performed in patients with recurrent GBM treated with bevacizumab and daily temozolomide. PET-MRI scans were performed prior to the first bevacizumab dose, 1 day after the first dose, and prior to the third dose of bevacizumab. We calculated tumor volume, vascular permeability (K trans), perfusion (cerebral blood flow), and the standardized uptake values (SUV) of [11C] temozolomide within the tumor. RESULTS: Twelve patients were enrolled, resulting in 23 evaluable scans. Within the entire contrast-enhancing tumor volume, both temozolomide uptake and vascular permeability decreased after initiation of bevacizumab in most patients, whereas change in perfusion was more variable. In subregions of the tumor where permeability was low and the blood-brain barrier not compromised, increased perfusion correlated with increased temozolomide uptake. CONCLUSIONS: Bevacizumab led to a decrease in permeability and concomitant delivery of temozolomide. However, in subregions of the tumor where permeability was low, increased perfusion improved delivery of temozolomide, suggesting that perfusion may modulate the delivery of chemotherapy in certain settings. These results support exploring whether lower doses of bevacizumab improve perfusion and concomitant drug delivery.

Machine Learning Models can Detect Aneurysm Rupture and Identify Clinical Features Associated with Rupture.

BACKGROUND: Machine learning (ML) has been increasingly used in medicine and neurosurgery. We sought to determine whether ML models can distinguish ruptured from unruptured aneurysms and identify features associated with rupture. METHODS: We performed a retrospective review of patients with intracranial aneurysms detected on vascular imaging at our institution between 2002 and 2018. The dataset was used to train 3 ML models (random forest, linear support vector machine [SVM], and radial basis function kernel SVM). Relative contributions of individual predictors were derived from the linear SVM model. RESULTS: Complete data were available for 845 aneurysms in 615 patients. Ruptured aneurysms (n = 309, 37%) were larger (mean 6.51 mm vs. 5.73 mm; P = 0.02) and more likely to be in the posterior circulation (20% vs. 11%; P < 0.001) than unruptured aneurysms. Area under the receiver operating curve was 0.77 for the linear SVM, 0.78 for the radial basis function kernel SVM models, and 0.81 for the random forest model. Aneurysm location and size were the 2 features that contributed most significantly to the model. Posterior communicating artery, anterior communicating artery, and posterior inferior cerebellar artery locations were most highly associated with rupture, whereas paraclinoid and middle cerebral artery locations had the strongest association with unruptured status. CONCLUSIONS: ML models are capable of accurately distinguishing ruptured from unruptured aneurysms and identifying features associated with rupture. Consistent with prior studies, location and size show the strongest association with aneurysm rupture.

Automatic assessment of glioma burden: a deep learning algorithm for fully automated volumetric and bidimensional measurement.

BACKGROUND: Longitudinal measurement of glioma burden with MRI is the basis for treatment response assessment. In this study, we developed a deep learning algorithm that automatically segments abnormal fluid attenuated inversion recovery (FLAIR) hyperintensity and contrast-enhancing tumor, quantitating tumor volumes as well as the product of maximum bidimensional diameters according to the Response Assessment in Neuro-Oncology (RANO) criteria (AutoRANO). METHODS: Two cohorts of patients were used for this study. One consisted of 843 preoperative MRIs from 843 patients with low- or high-grade gliomas from 4 institutions and the second consisted of 713 longitudinal postoperative MRI visits from 54 patients with newly diagnosed glioblastomas (each with 2 pretreatment "baseline" MRIs) from 1 institution. RESULTS: The automatically generated FLAIR hyperintensity volume, contrast-enhancing tumor volume, and AutoRANO were highly repeatable for the double-baseline visits, with an intraclass correlation coefficient (ICC) of 0.986, 0.991, and 0.977, respectively, on the cohort of postoperative GBM patients. Furthermore, there was high agreement between manually and automatically measured tumor volumes, with ICC values of 0.915, 0.924, and 0.965 for preoperative FLAIR hyperintensity, postoperative FLAIR hyperintensity, and postoperative contrast-enhancing tumor volumes, respectively. Lastly, the ICCs for comparing manually and automatically derived longitudinal changes in tumor burden were 0.917, 0.966, and 0.850 for FLAIR hyperintensity volume, contrast-enhancing tumor volume, and RANO measures, respectively. CONCLUSIONS: Our automated algorithm demonstrates potential utility for evaluating tumor burden in complex posttreatment settings, although further validation in multicenter clinical trials will be needed prior to widespread implementation.

Evaluating Multisite rCBV Consistency from DSC-MRI Imaging Protocols and Postprocessing Software Across the NCI Quantitative Imaging Network Sites Using a Digital Reference Object (DRO).

Relative cerebral blood volume (rCBV) cannot be used as a response metric in clinical trials, in part, because of variations in biomarker consistency and associated interpretation across sites, stemming from differences in image acquisition and postprocessing methods (PMs). This study leveraged a dynamic susceptibility contrast magnetic resonance imaging digital reference object to characterize rCBV consistency across 12 sites participating in the Quantitative Imaging Network (QIN), specifically focusing on differences in site-specific imaging protocols (IPs; n = 17), and PMs (n = 19) and differences due to site-specific IPs and PMs (n = 25). Thus, high agreement across sites occurs when 1 managing center processes rCBV despite slight variations in the IP. This result is most likely supported by current initiatives to standardize IPs. However, marked intersite disagreement was observed when site-specific software was applied for rCBV measurements. This study's results have important implications for comparing rCBV values across sites and trials, where variability in PMs could confound the comparison of therapeutic effectiveness and/or any attempts to establish thresholds for categorical response to therapy. To overcome these challenges and ensure the successful use of rCBV as a clinical trial biomarker, we recommend the establishment of qualifying and validating site- and trial-specific criteria for scanners and acquisition methods (eg, using a validated phantom) and the software tools used for dynamic susceptibility contrast magnetic resonance imaging analysis (eg, using a digital reference object where the ground truth is known).

ISLES 2016 and 2017-Benchmarking Ischemic Stroke Lesion Outcome Prediction Based on Multispectral MRI.

Performance of models highly depend not only on the used algorithm but also the data set it was applied to. This makes the comparison of newly developed tools to previously published approaches difficult. Either researchers need to implement others' algorithms first, to establish an adequate benchmark on their data, or a direct comparison of new and old techniques is infeasible. The Ischemic Stroke Lesion Segmentation (ISLES) challenge, which has ran now consecutively for 3 years, aims to address this problem of comparability. ISLES 2016 and 2017 focused on lesion outcome prediction after ischemic stroke: By providing a uniformly pre-processed data set, researchers from all over the world could apply their algorithm directly. A total of nine teams participated in ISLES 2015, and 15 teams participated in ISLES 2016. Their performance was evaluated in a fair and transparent way to identify the state-of-the-art among all submissions. Top ranked teams almost always employed deep learning tools, which were predominately convolutional neural networks (CNNs). Despite the great efforts, lesion outcome prediction persists challenging. The annotated data set remains publicly available and new approaches can be compared directly via the online evaluation system, serving as a continuing benchmark (www.isles-challenge.org).

Automated Diagnosis of Plus Disease in Retinopathy of Prematurity Using Deep Convolutional Neural Networks.

Importance: Retinopathy of prematurity (ROP) is a leading cause of childhood blindness worldwide. The decision to treat is primarily based on the presence of plus disease, defined as dilation and tortuosity of retinal vessels. However, clinical diagnosis of plus disease is highly subjective and variable. Objective: To implement and validate an algorithm based on deep learning to automatically diagnose plus disease from retinal photographs. Design, Setting, and Participants: A deep convolutional neural network was trained using a data set of 5511 retinal photographs. Each image was previously assigned a reference standard diagnosis (RSD) based on consensus of image grading by 3 experts and clinical diagnosis by 1 expert (ie, normal, pre-plus disease, or plus disease). The algorithm was evaluated by 5-fold cross-validation and tested on an independent set of 100 images. Images were collected from 8 academic institutions participating in the Imaging and Informatics in ROP (i-ROP) cohort study. The deep learning algorithm was tested against 8 ROP experts, each of whom had more than 10 years of clinical experience and more than 5 peer-reviewed publications about ROP. Data were collected from July 2011 to December 2016. Data were analyzed from December 2016 to September 2017. Exposures: A deep learning algorithm trained on retinal photographs. Main Outcomes and Measures: Receiver operating characteristic analysis was performed to evaluate performance of the algorithm against the RSD. Quadratic-weighted κ coefficients were calculated for ternary classification (ie, normal, pre-plus disease, and plus disease) to measure agreement with the RSD and 8 independent experts. Results: Of the 5511 included retinal photographs, 4535 (82.3%) were graded as normal, 805 (14.6%) as pre-plus disease, and 172 (3.1%) as plus disease, based on the RSD. Mean (SD) area under the receiver operating characteristic curve statistics were 0.94 (0.01) for the diagnosis of normal (vs pre-plus disease or plus disease) and 0.98 (0.01) for the diagnosis of plus disease (vs normal or pre-plus disease). For diagnosis of plus disease in an independent test set of 100 retinal images, the algorithm achieved a sensitivity of 93% with 94% specificity. For detection of pre-plus disease or worse, the sensitivity and specificity were 100% and 94%, respectively. On the same test set, the algorithm achieved a quadratic-weighted κ coefficient of 0.92 compared with the RSD, outperforming 6 of 8 ROP experts. Conclusions and Relevance: This fully automated algorithm diagnosed plus disease in ROP with comparable or better accuracy than human experts. This has potential applications in disease detection, monitoring, and prognosis in infants at risk of ROP.

Distributed deep learning networks among institutions for medical imaging.

Objective: Deep learning has become a promising approach for automated support for clinical diagnosis. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In this study, we propose methods of distributing deep learning models as an attractive alternative to sharing patient data. Methods: We simulate the distribution of deep learning models across 4 institutions using various training heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The training heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in 3 independent image collections (retinal fundus photos, mammography, and ImageNet). Results: We find that cyclical weight transfer resulted in a performance that was comparable to that of centrally hosted patient data. We also found that there is an improvement in the performance of cyclical weight transfer heuristic with a high frequency of weight transfer. Conclusions: We show that distributing deep learning models is an effective alternative to sharing patient data. This finding has implications for any collaborative deep learning study.

Semi-automated pulmonary nodule interval segmentation using the NLST data.

PURPOSE: To study the variability in volume change estimates of pulmonary nodules due to segmentation approaches used across several algorithms and to evaluate these effects on the ability to predict nodule malignancy. METHODS: We obtained 100 patient image datasets from the National Lung Screening Trial (NLST) that had a nodule detected on each of two consecutive low dose computed tomography (LDCT) scans, with an equal proportion of malignant and benign cases (50 malignant, 50 benign). Information about the nodule location for the cases was provided by a screen capture with a bounding box and its axial location was indicated. Five participating quantitative imaging network (QIN) institutions performed nodule segmentation using their preferred semi-automated algorithms with no manual correction; teams were allowed to provide additional manually corrected segmentations (analyzed separately). The teams were asked to provide segmentation masks for each nodule at both time points. From these masks, the volume was estimated for the nodule at each time point; the change in volume (absolute and percent change) across time points was estimated as well. We used the concordance correlation coefficient (CCC) to compare the similarity of computed nodule volumes (absolute and percent change) across algorithms. We used Logistic regression model on the change in volume (absolute change and percent change) of the nodules to predict the malignancy status, the area under the receiver operating characteristic curve (AUROC) and confidence intervals were reported. Because the size of nodules was expected to have a substantial effect on segmentation variability, analysis of change in volumes was stratified by lesion size, where lesions were grouped into those with a longest diameter of <8 mm and those with longest diameter ≥ 8 mm. RESULTS: We find that segmentation of the nodules shows substantial variability across algorithms, with the CCC ranging from 0.56 to 0.95 for change in volume (percent change in volume range was [0.15 to 0.86]) across the nodules. When examining nodules based on their longest diameter, we find the CCC had higher values for large nodules with a range of [0.54 to 0.93] among the algorithms, while percent change in volume was [0.3 to 0.95]. Compared to that of smaller nodules which had a range of [-0.0038 to 0.69] and percent change in volume was [-0.039 to 0.92]. The malignancy prediction results showed fairly consistent results across the institutions, the AUC using change in volume ranged from 0.65 to 0.89 (Percent change in volume was 0.64 to 0.86) for entire nodule range. Prediction improves for large nodule range (≥ 8 mm) with AUC range 0.75 to 0.90 (percent change in volume was 0.74 to 0.92). Compared to smaller nodule range (<8 mm) with AUC range 0.57 to 0.78 (percent change in volume was 0.59 to 0.77). CONCLUSIONS: We find there is a fairly high concordance in the size measurements for larger nodules (≥8 mm) than the lower sizes (<8 mm) across algorithms. We find the change in nodule volume (absolute and percent change) were consistent predictors of malignancy across institutions, despite using different segmentation algorithms. Using volume change estimates without corrections shows slightly lower predictability (for two teams).

Residual Convolutional Neural Network for the Determination of IDH Status in Low- and High-Grade Gliomas from MR Imaging.

Purpose: Isocitrate dehydrogenase (IDH) mutations in glioma patients confer longer survival and may guide treatment decision making. We aimed to predict the IDH status of gliomas from MR imaging by applying a residual convolutional neural network to preoperative radiographic data.Experimental Design: Preoperative imaging was acquired for 201 patients from the Hospital of University of Pennsylvania (HUP), 157 patients from Brigham and Women's Hospital (BWH), and 138 patients from The Cancer Imaging Archive (TCIA) and divided into training, validation, and testing sets. We trained a residual convolutional neural network for each MR sequence (FLAIR, T2, T1 precontrast, and T1 postcontrast) and built a predictive model from the outputs. To increase the size of the training set and prevent overfitting, we augmented the training set images by introducing random rotations, translations, flips, shearing, and zooming.Results: With our neural network model, we achieved IDH prediction accuracies of 82.8% (AUC = 0.90), 83.0% (AUC = 0.93), and 85.7% (AUC = 0.94) within training, validation, and testing sets, respectively. When age at diagnosis was incorporated into the model, the training, validation, and testing accuracies increased to 87.3% (AUC = 0.93), 87.6% (AUC = 0.95), and 89.1% (AUC = 0.95), respectively.Conclusions: We developed a deep learning technique to noninvasively predict IDH genotype in grade II-IV glioma using conventional MR imaging using a multi-institutional data set. Clin Cancer Res; 24(5); 1073-81. ©2017 AACR.