Intelligent noninvasive meningioma grading with a fully automatic segmentation using interpretable multiparametric deep learning.

OBJECTIVES: To establish a robust interpretable multiparametric deep learning (DL) model for automatic noninvasive grading of meningiomas along with segmentation. METHODS: In total, 257 patients with pathologically confirmed meningiomas (162 low-grade, 95 high-grade) who underwent a preoperative brain MRI, including T2-weighted (T2) and contrast-enhanced T1-weighted images (T1C), were included in the institutional training set. A two-stage DL grading model was constructed for segmentation and classification based on multiparametric three-dimensional U-net and ResNet. The models were validated in the external validation set consisting of 61 patients with meningiomas (46 low-grade, 15 high-grade). Relevance-weighted Class Activation Mapping (RCAM) method was used to interpret the DL features contributing to the prediction of the DL grading model. RESULTS: On external validation, the combined T1C and T2 model showed a Dice coefficient of 0.910 in segmentation and the highest performance for meningioma grading compared to the T2 or T1C only models, with an area under the curve (AUC) of 0.770 (95% confidence interval: 0.644-0.895) and accuracy, sensitivity, and specificity of 72.1%, 73.3%, and 71.7%, respectively. The AUC and accuracy of the combined DL grading model were higher than those of the human readers (AUCs of 0.675-0.690 and accuracies of 65.6-68.9%, respectively). The RCAM of the DL grading model showed activated maps at the surface regions of meningiomas indicating that the model recognized the features at the tumor margin for grading. CONCLUSIONS: An interpretable multiparametric DL model combining T1C and T2 can enable fully automatic grading of meningiomas along with segmentation. KEY POINTS: • The multiparametric DL model showed robustness in grading and segmentation on external validation. • The diagnostic performance of the combined DL grading model was higher than that of the human readers. • The RCAM interpreted that DL grading model recognized the meaningful features at the tumor margin for grading.

Opioid Prescriptions Among Women of Reproductive Age Enrolled in Medicaid - New York, 2008-2013.

Exposure to opioids during pregnancy can lead to adverse infant outcomes, including neonatal abstinence syndrome (1) and birth defects (2). Ascertaining opioid prescriptions for women who become pregnant or have no indication of contraceptive use is important to determine the number of women who are at potential risk for adverse fetal outcomes. The New York State (NYS) Department of Health (DOH) analyzed data for women aged 15-44 years (i.e., reproductive-aged women) enrolled in Medicaid to examine opioid drug prescriptions during 2008-2013. On the basis of Medicaid drug claims for any drug with an opioid ingredient, prescriptions were identified for the enrolled population of reproductive-aged women and for three subgroups: women whose diagnosis, procedure, and drug codes indicated contraceptive use or infertility; women who were not using contraceptives and not infertile; and women who had had a live birth during the reporting year. During 2008-2013, among all women of reproductive age, 20.0% received a prescription for a drug with an opioid component; the proportion was highest (27.3%) among women with an indication of contraceptive use or infertility, intermediate (17.3%) among women who had no indication of contraceptive use, and lowest (9.5%) among women who had had a live birth. Although New York's proportion of opioid prescriptions among female Medicaid recipients who had a live birth is lower than a recent U.S. estimate (3), these results suggest nearly one in 10 women in this group may have been exposed to opioids in the prenatal period.

Small Bowel Detection for Wireless Capsule Endoscopy Using Convolutional Neural Networks with Temporal Filtering.

By automatically classifying the stomach, small bowel, and colon, the reading time of the wireless capsule endoscopy (WCE) can be reduced. In addition, it is an essential first preprocessing step to localize the small bowel in order to apply automated small bowel lesion detection algorithms based on deep learning. The purpose of the study was to develop an automated small bowel detection method from long untrimmed videos captured from WCE. Through this, the stomach and colon can also be distinguished. The proposed method is based on a convolutional neural network (CNN) with a temporal filtering on the predicted probabilities from the CNN. For CNN, we use a ResNet50 model to classify three organs including stomach, small bowel, and colon. The hybrid temporal filter consisting of a Savitzky-Golay filter and a median filter is applied to the temporal probabilities for the "small bowel" class. After filtering, the small bowel and the other two organs are differentiated with thresholding. The study was conducted on dataset of 200 patients (100 normal and 100 abnormal WCE cases), which was divided into a training set of 140 cases, a validation set of 20 cases, and a test set of 40 cases. For the test set of 40 patients (20 normal and 20 abnormal WCE cases), the proposed method showed accuracy of 99.8% in binary classification for the small bowel. Transition time errors for gastrointestinal tracts were only 38.8 ± 25.8 seconds for the transition between stomach and small bowel and 32.0 ± 19.1 seconds for the transition between small bowel and colon, compared to the ground truth organ transition points marked by two experienced gastroenterologists.

Time trends in emergency department use among adults with intellectual and developmental disabilities.

BACKGROUND: Emergency Department (ED) visits are common among adults with intellectual and developmental disabilities (IDD). However, little is known about how ED use has varied over time in this population, or how it has been affected by recent Medicaid policy changes. OBJECTIVE: To examine temporal trends in ED use among adult Medicaid members with IDD in eight states that ranged in the extent to which they had implemented state-level Medicaid policy changes that might affect ED use. METHODS: We conducted repeated cross-sectional analyses of 2010-2016 Medicaid claims data. Quarterly analyses included adults ages 18-64 years with IDD (identified by diagnosis codes) who were continuously enrolled in Medicaid for the past 12 months. We assessed change in number of ED visits per 1000 member months from 2010 to 2016 overall and interacted with state level policy changes such as Medicaid expansion. RESULTS: States with no Medicaid expansion experienced an increase in ED visits (linear trend coefficient: 1.13, p < 0.01), while states operating expansion via waiver had a much smaller (non-significant) increase, and states with ACA-governed expansion had a decrease in ED visits (linear trend coefficient: 1.17, p < 0.01). Other policy changes had limited or no association with ED visits. CONCLUSIONS: Medicaid expansion was associated with modest reduction or limited increase in ED visits compared to no expansion. We found no consistent decrease in ED visits in association with other Medicaid policy changes.

Ultrathin crystalline-silicon-based strain gauges with deep learning algorithms for silent speech interfaces.

A wearable silent speech interface (SSI) is a promising platform that enables verbal communication without vocalization. The most widely studied methodology for SSI focuses on surface electromyography (sEMG). However, sEMG suffers from low scalability because of signal quality-related issues, including signal-to-noise ratio and interelectrode interference. Hence, here, we present a novel SSI by utilizing crystalline-silicon-based strain sensors combined with a 3D convolutional deep learning algorithm. Two perpendicularly placed strain gauges with minimized cell dimension (<0.1 mm2) could effectively capture the biaxial strain information with high reliability. We attached four strain sensors near the subject's mouths and collected strain data of unprecedently large wordsets (100 words), which our SSI can classify at a high accuracy rate (87.53%). Several analysis methods were demonstrated to verify the system's reliability, as well as the performance comparison with another SSI using sEMG electrodes with the same dimension, which exhibited a relatively low accuracy rate (42.60%).