Search

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.

Supplementary Data

We aimed to develop deep learning models using longitudinal chest X-rays (CXRs) and clinical data to predict in-hospital mortality of COVID-19 patients in the intensive care unit (ICU).Six hundred fifty-four patients (212 deceased, 442 alive, 5645 total CXRs) were identified across two institutions. Imaging and clinical data from one institution were used to train five longitudinal transformer-based networks applying five-fold cross-validation. The models were tested on data from the other institution, and pairwise comparisons were used to determine the best-performing models.A higher proportion of deceased patients had elevated white blood cell count, decreased absolute lymphocyte count, elevated creatine concentration, and incidence of cardiovascular and chronic kidney disease. A model based on pre-ICU CXRs achieved an AUC of 0.632 and an accuracy of 0.593, and a model based on ICU CXRs achieved an AUC of 0.697 and an accuracy of 0.657. A model based on all longitudinal CXRs (both pre-ICU and ICU) achieved an AUC of 0.702 and an accuracy of 0.694. A model based on clinical data alone achieved an AUC of 0.653 and an accuracy of 0.657. The addition of longitudinal imaging to clinical data in a combined model significantly improved performance, reaching an AUC of 0.727 (p = 0.039) and an accuracy of 0.732.The addition of longitudinal CXRs to clinical data significantly improves mortality prediction with deep learning for COVID-19 patients in the ICU.• Deep learning was used to predict mortality in COVID-19 ICU patients. • Serial radiographs and clinical data were used. • The models could inform clinical decision-making and resource allocation.

In the era of data-driven medicine, rapid access and accurate interpretation of medical images are becoming increasingly important. The DICOM Image ANalysis and Archive (DIANA) system is an open-source, lightweight, and scalable Python interface that enables users to interact with hospital Picture Archiving and Communications Systems (PACS) to access such data. In this work, DIANA functionality was detailed and evaluated in the context of retrospective PACS data retrieval and two prospective clinical artificial intelligence (AI) pipelines: bone age (BA) estimation and intra-cranial hemorrhage (ICH) detection. DIANA orchestrates activity beginning with post-acquisition study discovery and ending with online notifications of findings. For AI applications, system latency (exam completion to system report time) was quantified and compared to that of clinicians (exam completion to initial report creation time). Mean DIANA latency was 9.04 ± 3.83 and 20.17 ± 10.16 min compared to clinician latency of 51.52 ± 58.9 and 65.62 ± 110.39 min for BA and ICH, respectively, with DIANA latencies being significantly lower (p

Artificial intelligence (AI) has illuminated a clear path towards an evolving health-care system replete with enhanced precision and computing capabilities. Medical imaging analysis can be strengthened by machine learning as the multidimensional data generated by imaging naturally lends itself to hierarchical classification. In this Review, we describe the role of machine intelligence in image-based endocrine cancer diagnostics. We first provide a brief overview of AI and consider its intuitive incorporation into the clinical workflow. We then discuss how AI can be applied for the characterization of adrenal, pancreatic, pituitary and thyroid masses in order to support clinicians in their diagnostic interpretations. This Review also puts forth a number of key evaluation criteria for machine learning in medicine that physicians can use in their appraisals of these algorithms. We identify mitigation strategies to address ongoing challenges around data availability and model interpretability in the context of endocrine cancer diagnosis. Finally, we delve into frontiers in systems integration for AI, discussing automated pipelines and evolving computing platforms that leverage distributed, decentralized and quantum techniques., This Review explains core concepts in artificial intelligence (AI) and machine learning for endocrinologists. AI applications in endocrine cancer diagnostics are highlighted as well as research challenges and future directions for the field., Key points Developments in machine intelligence have been made possible by the increase in data ubiquity and computing power and have the potential to enhance image segmentation, analysis and workflow in non-invasive endocrine cancer diagnostics.Improved adherence to consensus reporting standards and evaluation criteria in artificial intelligence (AI) for medical image analysis is urgently needed in the field of endocrine cancer diagnostics as this will enable meaningful cross-study comparison.A centralized inventory to track diagnostic algorithms in oncologic endocrinology that are in active clinical use would improve performance auditing and algorithm stewardship.The looming risk of excessive intervention in endocrine cancers can be addressed with the improved detection facilitated by AI, possibly via correlation with prognostic data for improved risk stratification.Poor data availability continues to stymie the development of robust machine learning applications, particularly in rare endocrine cancers; solutions to this problem might include database curation, pre-training techniques and workflow automation.Other breakthroughs in machine intelligence will come with the exploration of alternative computing frameworks, such as decentralized, distributed and quantum networks, that might enhance model training and efficiency.

In this study, we developed radiomic models that utilize a combination of imaging features and clinical variables to distinguish endometrial cancer (EC) from normal endometrium on routine computed tomography (CT). A total of 926 patients consisting of 416 endometrial cancer (EC) and 510 normal endometrium were included. The CT images of these patients were segmented manually, and divided into training, validation, testing and external testing sets. Non-texture and texture features of these images with endometrium or uterus as region of interest were extracted. The clinical feature “age” was also included in the feature set. Feature selection and machine learning classifier were applied to normalized feature set. This manual optimized combination was then compared with the best pipeline exported by Tree-Based Pipeline Optimization Tool (TPOT) on testing and external testing set. The performances of these machine learning pipelines were compared to that of radiologists. The manual expert optimized pipeline using the “reliefF” feature selection method and “Bagging” classifier on the external testing set achieved a test ROC AUC of 0.73, accuracy of 0.73 (95% CI 0.62–0.82), sensitivity of 0.64 (95% CI 0.45–0.79), and specificity of 0.78 (95% CI 0.65–0.87), while TPOT achieved a test ROC AUC of 0.79, accuracy of 0.80 (95% CI 0.70–0.87), sensitivity of 0.61 (95% CI 0.43–0.77), and specificity of 0.90 (95% CI 0.78–0.96). When compared to average radiologist performance, the TPOT achieved higher test accuracy (0.80 vs. 0.49, p

Objectives Early recognition of coronavirus disease 2019 (COVID-19) severity can guide patient management. However, it is challenging to predict when COVID-19 patients will progress to critical illness. This study aimed to develop an artificial intelligence system to predict future deterioration to critical illness in COVID-19 patients. Methods An artificial intelligence (AI) system in a time-to-event analysis framework was developed to integrate chest CT and clinical data for risk prediction of future deterioration to critical illness in patients with COVID-19. Results A multi-institutional international cohort of 1,051 patients with RT-PCR confirmed COVID-19 and chest CT was included in this study. Of them, 282 patients developed critical illness, which was defined as requiring ICU admission and/or mechanical ventilation and/or reaching death during their hospital stay. The AI system achieved a C-index of 0.80 for predicting individual COVID-19 patients’ to critical illness. The AI system successfully stratified the patients into high-risk and low-risk groups with distinct progression risks (p < 0.0001). Conclusions Using CT imaging and clinical data, the AI system successfully predicted time to critical illness for individual patients and identified patients with high risk. AI has the potential to accurately triage patients and facilitate personalized treatment. Key Point • AI system can predict time to critical illness for patients with COVID-19 by using CT imaging and clinical data. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-021-08049-8.

In this study, we developed a deep learning pipeline that detects large vessel occlusion (LVO) and predicts functional outcome based on computed tomography angiography (CTA) images to improve the management of the LVO patients.A series identifier picked out 8650 LVO-protocoled studies from 2015 to 2019 at Rhode Island Hospital with an identified thin axial series that served as the data pool. Data were annotated into 2 classes: 1021 LVOs and 7629 normal. The Inception-V1 I3D architecture was applied for LVO detection. For outcome prediction, 323 patients undergoing thrombectomy were selected. A 3D convolution neural network (CNN) was used for outcome prediction (30-day mRS) with CTA volumes and embedded pre-treatment variables as inputs.For LVO-detection model, CTAs from 8,650 patients (median age 68 years, interquartile range (IQR): 58-81; 3934 females) were analyzed. The cross-validated AUC for LVO vs. not was 0.74 (95% CI: 0.72-0.75). For the mRS classification model, CTAs from 323 patients (median age 75 years, IQR: 63-84; 164 females) were analyzed. The algorithm achieved a test AUC of 0.82 (95% CI: 0.79-0.84), sensitivity of 89%, and specificity 66%. The two models were then integrated with hospital infrastructure where CTA was collected in real-time and processed by the model. If LVO was detected, interventionists were notified and provided with predicted clinical outcome information.3D CNNs based on CTA were effective in selecting LVO and predicting LVO mechanical thrombectomy short-term prognosis. End-to-end AI platform allows users to receive immediate prognosis prediction and facilitates clinical workflow.

Background Longitudinal measurement of tumor burden with magnetic resonance imaging (MRI) is an essential component of response assessment in pediatric brain tumors. We developed a fully automated pipeline for the segmentation of tumors in pediatric high-grade gliomas, medulloblastomas, and leptomeningeal seeding tumors. We further developed an algorithm for automatic 2D and volumetric size measurement of tumors. Methods The preoperative and postoperative cohorts were randomly split into training and testing sets in a 4:1 ratio. A 3D U-Net neural network was trained to automatically segment the tumor on T1 contrast-enhanced and T2/FLAIR images. The product of the maximum bidimensional diameters according to the RAPNO (Response Assessment in Pediatric Neuro-Oncology) criteria (AutoRAPNO) was determined. Performance was compared to that of 2 expert human raters who performed assessments independently. Volumetric measurements of predicted and expert segmentations were computationally derived and compared. Results A total of 794 preoperative MRIs from 794 patients and 1003 postoperative MRIs from 122 patients were included. There was excellent agreement of volumes between preoperative and postoperative predicted and manual segmentations, with intraclass correlation coefficients (ICCs) of 0.912 and 0.960 for the 2 preoperative and 0.947 and 0.896 for the 2 postoperative models. There was high agreement between AutoRAPNO scores on predicted segmentations and manually calculated scores based on manual segmentations (Rater 2 ICC = 0.909; Rater 3 ICC = 0.851). Lastly, the performance of AutoRAPNO was superior in repeatability to that of human raters for MRIs with multiple lesions. Conclusions Our automated deep learning pipeline demonstrates potential utility for response assessment in pediatric brain tumors. The tool should be further validated in prospective studies.

Aims To determine if neurologic symptoms at admission can predict adverse outcomes in patients with severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2). Methods Electronic medical records of 1053 consecutively hospitalized patients with laboratory‐confirmed infection of SARS‐CoV‐2 from one large medical center in the USA were retrospectively analyzed. Univariable and multivariable Cox regression analyses were performed with the calculation of areas under the curve (AUC) and concordance index (C‐index). Patients were stratified into subgroups based on the presence of encephalopathy and its severity using survival statistics. In sensitivity analyses, patients with mild/moderate and severe encephalopathy (defined as coma) were separately considered. Results Of 1053 patients (mean age 52.4 years, 48.0% men [n = 505]), 35.1% (n = 370) had neurologic manifestations at admission, including 10.3% (n = 108) with encephalopathy. Encephalopathy was an independent predictor for death (hazard ratio [HR] 2.617, 95% confidence interval [CI] 1.481–4.625) in multivariable Cox regression. The addition of encephalopathy to multivariable models comprising other predictors for adverse outcomes increased AUCs (mortality: 0.84–0.86, ventilation/ intensive care unit [ICU]: 0.76–0.78) and C‐index (mortality: 0.78 to 0.81, ventilation/ICU: 0.85–0.86). In sensitivity analyses, risk stratification survival curves for mortality and ventilation/ICU based on severe encephalopathy (n = 15) versus mild/moderate encephalopathy (n = 93) versus no encephalopathy (n = 945) at admission were discriminative (p, Patients with severe acute respiratory syndrome coronavirus 2 infection have a high prevalence of neurologic manifestations, including headache, encephalopathy, dizziness, taste, and smell impairment. Patients with encephalopathy at admission predict later progression to death and mechanical ventilation/ICU admission in SARS‐CoV‐2 infection, which may have important implications for risk stratification in clinical practice.

Infection with SARS-CoV-2 ranges from an asymptomatic condition to a severe and sometimes fatal disease, with mortality most frequently being the result of acute lung injury. The role of imaging has evolved during the pandemic, with CT initially being an alternative and possibly superior testing method compared with reverse transcriptase-polymerase chain reaction (RT-PCR) testing and evolving to having a more limited role based on specific indications. Several classification and reporting schemes were developed for chest imaging early during the pandemic for patients suspected of having COVID-19 to aid in triage when the availability of RT-PCR testing was limited and its level of performance was unclear. Interobserver agreement for categories with findings typical of COVID-19 and those suggesting an alternative diagnosis is high across multiple studies. Furthermore, some studies looking at the extent of lung involvement on chest radiographs and CT images showed correlations with critical illness and a need for mechanical ventilation. In addition to pulmonary manifestations, cardiovascular complications such as thromboembolism and myocarditis have been ascribed to COVID-19, sometimes contributing to neurologic and abdominal manifestations. Finally, artificial intelligence has shown promise for use in determining both the diagnosis and prognosis of COVID-19 pneumonia with respect to both radiography and CT.

ObjectiveThe objective of this study is to examine the current landscape of FDA-approved AI medical imaging devices and identify trends in clinical validation strategy.Materials and MethodsWe conducted a retrospective study that analyzed data extracted from the American College of Radiology (ACR) Data Science Institute AI Central database as of November 2021 to identify trends in FDA clearance of AI products related to medical imaging. Product and clinical validation information of each device was gathered from their respective public 510(k) summary or de novo request submission, depending on their type of authorization.ResultsOverall, the database included a total of 151 AI algorithms that were cleared by the FDA between 2008 and November 2021. Out of the 151 FDA summaries reviewed, 97 (64.2%) reported the use of clinical data to validate their device. Of these 151 summaries, 81 (53.6%) reported the total number of patient cases used during validation, with the average number of cases being 799 (SD: 1363) and the range of cases spanning from 15 to 9122. A total of 51 (33.8%) AI devices characterized their clinical data as multicenter, 3 (2.0%) as single-center, and the remaining 97 (64.2%) did not specify. The ground truth used for clinical validation was specified in 78 (51.6%) FDA summaries.Discussion and ConclusionA wide breadth of AI algorithms have been developed for medical imaging. Most of the devices’ FDA summaries mention their use of clinical data and patient cases for device validation, emphasizing their utility in real clinical practice.

Clear cell renal cell carcinoma (ccRCC) is the most common subtype of renal cell carcinoma. Currently, there is a lack of noninvasive methods to stratify ccRCC prognosis prior to any invasive therapies. The purpose of this study was to preoperatively predict the tumor stage, size, grade, and necrosis (SSIGN) score of ccRCC using MRI-based radiomics. A multicenter cohort of 364 histopathologically confirmed ccRCC patients (272 low [

There currently lacks a noninvasive and accurate method to distinguish benign and malignant ovarian lesion prior to treatment. This study developed a deep learning algorithm that distinguishes benign from malignant ovarian lesion by applying a convolutional neural network on routine MR imaging. Five hundred forty-five lesions (379 benign and 166 malignant) from 451 patients from a single institution were divided into training, validation, and testing set in a 7:2:1 ratio. Model performance was compared with four junior and three senior radiologists on the test set. Compared with junior radiologists averaged, the final ensemble model combining MR imaging and clinical variables had a higher test accuracy (0.87 vs 0.64, p

Key Points Question Which applications of artificial intelligence (AI) in biomedical research could generate the most translational value (ie, ability to lead to development that has measurable value for human health)? Findings In this cohort study using bibliometric analysis, 2 citation-based metrics were used as proxies for translational impact, and 75 unique medical applications of AI were identified from a set of 16 629 National Institutes of Health awards related to AI, using natural language processing. Varied applications were found to have high translatability, but bias against applications in basic biochemical analysis was evident. Meaning These findings suggest that National Institutes of Health award data applications of AI in biomedical research are not equivalent: some demonstrate greater potential for translational impact than others., Importance Despite the rapid growth of interest and diversity in applications of artificial intelligence (AI) to biomedical research, there are limited objective ways to characterize the potential for use of AI in clinical practice. Objective To examine what types of medical AI have the greatest estimated translational impact (ie, ability to lead to development that has measurable value for human health) potential. Design, Setting, and Participants In this cohort study, research grants related to AI awarded between January 1, 1985, and December 31, 2020, were identified from a National Institutes of Health (NIH) award database. The text content for each award was entered into a Natural Language Processing (NLP) clustering algorithm. An NIH database was also used to extract citation data, including the number of citations and approximate potential to translate (APT) score for published articles associated with the granted awards to create proxies for translatability. Exposures Unsupervised assignment of AI-related research awards to application topics using NLP. Main Outcomes and Measures Annualized citations per $1 million funding (ACOF) and average APT score for award-associated articles, grouped by application topic. The APT score is a machine-learning based metric created by the NIH Office of Portfolio Analysis that quantifies the likelihood of future citation by a clinical article. Results A total of 16 629 NIH awards related to AI were included in the analysis, and 75 applications of AI were identified. Total annual funding for AI grew from $17.4 million in 1985 to $1.43 billion in 2020. By average APT, interpersonal communication technologies (0.488; 95% CI, 0.472-0.504) and population genetics (0.463; 95% CI, 0.453-0.472) had the highest translatability; environmental health (ACOF, 1038) and applications focused on the electronic health record (ACOF, 489) also had high translatability. The category of applications related to biochemical analysis was found to have low translatability by both metrics (average APT, 0.393; 95% CI, 0.388-0.398; ACOF, 246). Conclusions and Relevance Based on this study's findings, data on grants from the NIH can apparently be used to identify and characterize medical applications of AI to understand changes in academic productivity, funding support, and potential for translational impact. This method may be extended to characterize other research domains., This cohort study examines the translatability of findings from studies on artificial intelligence funded by the National Institutes of Health (NIH).

Background Coronavirus disease 2019 (COVID-19) and pneumonia of other diseases share similar CT characteristics, which contributes to the challenges in differentiating them with high accuracy. Purpose To establish and evaluate an artificial intelligence (AI) system for differentiating COVID-19 and other pneumonia at chest CT and assessing radiologist performance without and with AI assistance. Materials and Methods A total of 521 patients with positive reverse transcription polymerase chain reaction results for COVID-19 and abnormal chest CT findings were retrospectively identified from 10 hospitals from January 2020 to April 2020. A total of 665 patients with non-COVID-19 pneumonia and definite evidence of pneumonia at chest CT were retrospectively selected from three hospitals between 2017 and 2019. To classify COVID-19 versus other pneumonia for each patient, abnormal CT slices were input into the EfficientNet B4 deep neural network architecture after lung segmentation, followed by a two-layer fully connected neural network to pool slices together. The final cohort of 1186 patients (132 583 CT slices) was divided into training, validation, and test sets in a 7:2:1 and equal ratio. Independent testing was performed by evaluating model performance in separate hospitals. Studies were blindly reviewed by six radiologists without and then with AI assistance. Results The final model achieved a test accuracy of 96% (95% confidence interval [CI]: 90%, 98%), a sensitivity of 95% (95% CI: 83%, 100%), and a specificity of 96% (95% CI: 88%, 99%) with area under the receiver operating characteristic curve of 0.95 and area under the precision-recall curve of 0.90. On independent testing, this model achieved an accuracy of 87% (95% CI: 82%, 90%), a sensitivity of 89% (95% CI: 81%, 94%), and a specificity of 86% (95% CI: 80%, 90%) with area under the receiver operating characteristic curve of 0.90 and area under the precision-recall curve of 0.87. Assisted by the probabilities of the model, the radiologists achieved a higher average test accuracy (90% vs 85%, Δ = 5

Background Despite its high sensitivity in diagnosing COVID-19 in a screening population, chest CT appearances of COVID 19 pneumonia are thought to be non-specific. Purpose To assess the performance of United States (U.S.) and Chinese radiologists in differentiating COVID-19 from viral pneumonia on chest CT. Methods A total of 219 patients with both positive COVID-19 by RT-PCR and abnormal chest CT findings were retrospectively identified from 7 Chinese hospitals in Hunan Providence, China from January 6 to February 20, 2020. A total of 205 patients with positive Respiratory Pathogen Panel for viral pneumonia and CT findings consistent with or highly suspicious for pneumonia by original radiology interpretation within 7 days of each other were identified from Rhode Island Hospital in Providence, RI. Three Chinese radiologists blindly reviewed all chest CTs (n=424) to differentiate COVID-19 from viral pneumonia. A sample of 58 age-matched cases was randomly selected and evaluated by 4 U.S. radiologists in a similar fashion. Different CT features were recorded and compared between the two groups. Results For all chest CTs, three Chinese radiologists correctly differentiated COVID-19 from non-COVID-19 pneumonia 83% (350/424), 80% (338/424), and 60% (255/424) of the time, respectively. The seven radiologists had sensitivities of 80%, 67%, 97%, 93%, 83%, 73% and 70% and specificities of 100%, 93%, 7%, 100%, 93%, 93%, 100%. Compared to non-COVID-19 pneumonia, COVID-19 pneumonia was more likely to have a peripheral distribution (80% vs. 57%, p

The ability to reliably distinguish benign from malignant solid liver lesions on ultrasonography can increase access, decrease costs, and help to better triage patients for biopsy. In this study, we used deep learning to differentiate benign from malignant focal solid liver lesions based on their ultrasound appearance. Among the 596 patients who met the inclusion criteria, there were 911 images of individual liver lesions, of which 535 were malignant and 376 were benign. Our training set contained 660 lesions augmented dynamically during training for a total of 330,000 images; our test set contained 79 images. A neural network with ResNet50 architecture was fine-tuned using pre-trained weights on ImageNet. Non-cystic liver lesions with definite diagnosis by histopathology or MRI were included. Accuracy of the final model was compared with expert interpretation. Two separate datasets were used in training and evaluation, one with all lesions and one with lesions deemed to be of uncertain diagnosis based on the Code Abdomen rating system. Our model trained on the complete set of all lesions achieved a test accuracy of 0.84 (95% CI 0.74–0.90) compared to expert 1 with a test accuracy of 0.80 (95% CI 0.70–0.87) and expert 2 with a test accuracy of 0.73 (95% CI 0.63–0.82). Our model trained on the uncertain set of lesions achieved a test accuracy of 0.79 (95% CI 0.69–0.87) compared to expert 1 with a test accuracy of 0.70 (95% CI 0.59–0.78) and expert 2 with a test accuracy of 0.66 (95% CI 0.55–0.75). On the uncertain dataset, compared to all experts averaged, the model had higher test accuracy (0.79 vs. 0.68, p = 0.025). Deep learning algorithms proposed in the current study improve differentiation of benign from malignant ultrasound-captured solid liver lesions and perform comparably to expert radiologists. Deep learning tools can potentially be used to improve the accuracy and efficiency of clinical workflows.

BACKGROUND AND PURPOSE: Differentiating the types of pediatric posterior fossa tumors on routine imaging may help in preoperative evaluation and guide surgical resection planning. However, qualitative radiologic MR imaging review has limited performance. This study aimed to compare different machine learning approaches to classify pediatric posterior fossa tumors on routine MR imaging. MATERIALS AND METHODS: This retrospective study included preoperative MR imaging of 288 patients with pediatric posterior fossa tumors, including medulloblastoma (n = 111), ependymoma (n = 70), and pilocytic astrocytoma (n = 107). Radiomics features were extracted from T2-weighted images, contrast-enhanced T1-weighted images, and ADC maps. Models generated by standard manual optimization by a machine learning expert were compared with automatic machine learning via the Tree-Based Pipeline Optimization Tool for performance evaluation. RESULTS: For 3-way classification, the radiomics model by automatic machine learning with the Tree-Based Pipeline Optimization Tool achieved a test micro-averaged area under the curve of 0.91 with an accuracy of 0.83, while the most optimized model based on the feature-selection method χ(2) score and the Generalized Linear Model classifier achieved a test micro-averaged area under the curve of 0.92 with an accuracy of 0.74. Tree-Based Pipeline Optimization Tool models achieved significantly higher accuracy than average qualitative expert MR imaging review (0.83 versus 0.54, P

OBJECTIVE The aim of the study was to evaluate efficacy of chemoembolization and radioembolization in treating patients with recurrent hypoglycemia secondary to metastatic insulinoma. METHODS A retrospective review was performed of all patients with metastatic insulinoma treated with liver-directed therapy (LDT) at a large academic medical center from January 1998 to August 2017. Primary outcomes included blood glucose levels, occurrence of symptomatic hypoglycemic episodes, and tumor imaging response rates. RESULTS Seven patients were identified (4 male patients). The mean age at the first LDT was 60.9 (standard deviation [SD], 9.2) years. The median follow-up was 1.8 years. Thirty-three sessions of LDT were performed including 30 sessions of chemoembolization and 3 sessions of radioembolization. Technical success rate was 97% (32/33 procedures) with an initial clinical success rate of 100%, defined as absence of recurrent symptomatic hypoglycemia within 1 month after first cycle of LDT and overall clinical success rate of 85%. Random daytime glucose levels increased from 75.0 (SD, 26) mg/dL to 152.8 (SD, 52.4) mg/dL after LDT. The mean time to recurrence of intractable hypoglycemia was 21 (SD, 9) months. No severe complications were reported. CONCLUSIONS Liver-directed therapy with chemoembolization or radioembolization is effective for treatment of symptomatic hypoglycemia secondary to malignant insulinoma metastatic to liver.

Purpose: With increasing incidence of renal mass, it is important to make a pretreatment differentiation between benign renal mass and malignant tumor. We aimed to develop a deep learning model that distinguishes benign renal tumors from renal cell carcinoma (RCC) by applying a residual convolutional neural network (ResNet) on routine MR imaging. Experimental Design: Preoperative MR images (T2-weighted and T1-postcontrast sequences) of 1,162 renal lesions definitely diagnosed on pathology or imaging in a multicenter cohort were divided into training, validation, and test sets (70:20:10 split). An ensemble model based on ResNet was built combining clinical variables and T1C and T2WI MR images using a bagging classifier to predict renal tumor pathology. Final model performance was compared with expert interpretation and the most optimized radiomics model. Results: Among the 1,162 renal lesions, 655 were malignant and 507 were benign. Compared with a baseline zero rule algorithm, the ensemble deep learning model had a statistically significant higher test accuracy (0.70 vs. 0.56, P = 0.004). Compared with all experts averaged, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.60, P = 0.053), sensitivity (0.92 vs. 0.80, P = 0.017), and specificity (0.41 vs. 0.35, P = 0.450). Compared with the radiomics model, the ensemble deep learning model had higher test accuracy (0.70 vs. 0.62, P = 0.081), sensitivity (0.92 vs. 0.79, P = 0.012), and specificity (0.41 vs. 0.39, P = 0.770). Conclusions: Deep learning can noninvasively distinguish benign renal tumors from RCC using conventional MR imaging in a multi-institutional dataset with good accuracy, sensitivity, and specificity comparable with experts and radiomics.

Background The benefit of adjuvant radiotherapy (RT) is uncertain in elderly patients diagnosed with World Health Organization (WHO) grade III meningiomas. Methods The US National Cancer Database was used to identify elderly patients (aged over 60) with histologically confirmed WHO grade III meningiomas who underwent surgery with or without adjuvant external beam RT. Overall survival (OS) was evaluated by Kaplan-Meier analysis, multivariable Cox proportional hazard models, and propensity score–matched analysis. Results Among the 254 patients who met inclusion criteria, 151 (58.4%) underwent surgery followed by RT and 103 (41.6%) surgery without RT. The 5-year relative survival rate was 57.8% (95% confidence interval [CI], 38.7%–85.4%) in the adjuvant RT group and 38.1% (95% CI, 21.9%–55.9%) in the group without RT when compared with age- and gender-matched US population. Adjuvant RT was associated with longer OS on both univariate (P = 0.011) and multivariable analyses (adjusted hazard ratio [AHR], 0.663; 95% CI, 0.449–0.978; P = 0.038). However, adjuvant RT did not significantly improve OS in a propensity score–matched analysis (P = 0.271) after matching the 2 cohorts on age, race, comorbidity, extent of resection, and tumor size. In patients who underwent subtotal resection or biopsy (n = 106), adjuvant RT significantly benefited OS on univariate analysis (P = 0.023), but did not show statistically significant benefit on multivariate analysis (AHR, 0.594; 95% CI, 0.337–1.046; P = 0.071). Adjuvant RT was not associated with improved OS in the elderly patients who underwent gross total resection (AHR, 0.764; 95% CI, 0.434–1.344; P = 0.350). Conclusion Adjuvant external beam RT may not provide any survival benefit for elderly patients with WHO grade III meningioma after gross total resection.