Mean Area Under The Curve Values Research Articles

Prediction of MicroRNA-Disease Potential Association Based on Sparse Learning and Multilayer Random Walks.

More and more studies have shown that microRNAs (miRNAs) play an indispensable role in the study of complex diseases in humans. Traditional biological experiments to detect miRNA-disease associations are expensive and time-consuming. Therefore, it is necessary to propose efficient and meaningful computational models to predict miRNA-disease associations. In this study, we aim to propose a miRNA-disease association prediction model based on sparse learning and multilayer random walks (SLMRWMDA). The miRNA-disease association matrix is decomposed and reconstructed by the sparse learning method to obtain richer association information, and at the same time, the initial probability matrix for the random walk with restart algorithm is obtained. The disease similarity network, miRNA similarity network, and miRNA-disease association network are used to construct heterogeneous networks, and the stable probability is obtained based on the topological structure features of diseases and miRNAs through a multilayer random walk algorithm to predict miRNA-disease potential association. The experimental results show that the prediction accuracy of this model is significantly improved compared with the previous related models. We evaluated the model using global leave-one-out cross-validation (global LOOCV) and fivefold cross-validation (5-fold CV). The area under the curve (AUC) value for the LOOCV is 0.9368. The mean AUC value for 5-fold CV is 0.9335 and the variance is 0.0004. In the case study, the results show that SLMRWMDA is effective in inferring the potential association of miRNA-disease.

Machine learning prediction models for in-hospital postoperative functional outcome after moderate-to-severe traumatic brain injury.

This study aims to utilize machine learning (ML) and logistic regression (LR) models to predict surgical outcomes among patients with traumatic brain injury (TBI) based on admission examination, assisting in making optimal surgical treatment decision for these patients. We conducted a retrospective review of patients hospitalized in our department for moderate-to-severe TBI. Patients admitted between October 2011 and October 2022 were assigned to the training set, while patients admitted between November 2022 and May 2023 were designated as the external validation set. Five ML algorithms and LR model were employed to predict the postoperative Glasgow Outcome Scale (GOS) status at discharge using clinical and routine blood data collected upon admission. The Shapley (SHAP) plot was utilized for interpreting the models. A total of 416 patients were included in this study, and they were divided into the training set (n = 396) and the external validation set (n = 47). The ML models, using both clinical and routine blood data, were able to predict postoperative GOS outcomes with area under the curve (AUC) values ranging from 0.860 to 0.900 during the internal cross-validation and from 0.801 to 0.890 during the external validation. In contrast, the LR model had the lowest AUC values during the internal and external validation (0.844 and 0.567, respectively). When blood data was not available, the ML models achieved AUCs of 0.849 to 0.870 during the internal cross-validation and 0.714 to 0.861 during the external validation. Similarly, the LR model had the lowest AUC values (0.821 and 0.638, respectively). Through repeated cross-validation analysis, we found that routine blood data had a significant association with higher mean AUC values in all ML and LR models. The SHAP plot was used to visualize the contributions of all predictors and highlighted the significance of blood data in the lightGBM model. The study concluded that ML models could provide rapid and accurate predictions for postoperative GOS outcomes at discharge following moderate-to-severe TBI. The study also highlighted the crucial role of routine blood tests in improving such predictions, and may contribute to the optimization of surgical treatment decision-making for patients with TBI.

Machine learning analysis of pathological images to predict 1-year progression-free survival of immunotherapy in patients with small-cell lung cancer

BackgroundIn small-cell lung cancer (SCLC), the tumor immune microenvironment (TIME) could be a promising biomarker for immunotherapy, but objectively evaluating TIME remains challenging. Hence, we aimed to develop a predictive...

Geriatric Nutritional Risk Index Combined with Calf Circumference Can be a Good Predictor of Prognosis in Patients Undergoing Surgery for Gastric or Colorectal Cancer.

To explore the effect of combined hematological and physical measurement indicators on the prognosis of patients undergoing surgery for gastric or colorectal cancer and to screen for the best prognostic indicators. Gastric and colorectal cancer is a widespread health concern worldwide and one of the major contributors to cancer-related death. The hematological and physical measurement indicators have been shown to associate with the prognosis of patients undergoing surgery for gastric or colorectal cancer, respectively, but it is still unclear whether the combination of the two can reflect the prognosis more effectively. Thirteen hematological indicators and 5 physical measurement indicators were selected in this study, and the most promising ones were screened using LASSO regression. Then, the best prognostic indicators were selected by time-ROC curves. Survival curves were constructed using the Kaplan-Meier method, and the effects of hematological and physical measurement indicators on the prognosis of patients undergoing surgery for gastric or colorectal cancers were evaluated by Cox proportional risk regression analysis. In addition, the relationship between hematological and physical measurement indicators on secondary outcomes, including length of stay, hospitalization costs, intensive care unit (ICU) admission, and patients' subjective global assessment scores (PGSGA), was explored. After initial screening, among the hematological indicators, the geriatric nutritional risk index (GNRI) showed the highest mean area under the curve (AUC) values. Among body measures, calf circumference (CC) showed the highest mean AUC value. Further analyses showed that the combination of combined nutritional prognostic index (GNRI) and calf circumference (CC) (GNRI-CC) had the best performance in predicting the prognosis of patients undergoing surgery for gastric or colorectal cancers. Low GNRI, low CC, and low GNRI-low CC increased the risk of death by 44%, 48%, and 104%, respectively. Sensitivity analyses showed the same trend. In addition, low GNRI-low CC increased the risk of malnutrition by 17%. This study emphasizes that a combination of blood measures and body measures is essential to accurately assess the prognosis of patients undergoing surgery for gastric or colorectal cancers. The GNRI-CC is a good prognostic indicator and can also assess the risk of possible malnutrition.

Cerebrolysin provides effective protection on high glucose-induced neuropathy in cultured rat dorsal root ganglion neurons

Cerebrolysin, an endogenous peptide with neuroprotective and neurotrophic properties, indicated to be beneficial on diabetic neuropathy by preliminary clinical and experimental studies but without evidence on central or peripheral action. Dorsal root ganglion (DRG) neurons, based on involvement of pain sensation in both health and disease as first relay centers for transmission and processing of peripheral nociceptive sensory signals, was used to investigate possible effects of Cerebrolysin on high glucose-induced neuropathy, as model. DRG’s were obtained from adult rats and the isolated neurons were seeded on E-Plate®’s equipped with gold microelectrodes, and incubated in culture media in a CO2 incubator at 37 C. DRGs were exposed to high glucose (50 mM) in the absence and presence of different concentrations of Cerebrolysin ® (2–40 mg/ml). Cell index (derived from cell viability and neurite outgrowth) was recorded with Real-Time Cell Analyzer and was used as primary outcome measure. High glucose-induced cellular neuropathy and neuroprotective effects of Cerebrolysin was evaluated from area under the curve (AUC) of cell index-time graphs. Exposure of DRG neurons to high glucose caused a rapid and persistent decrease in the mean AUC values compared to normoglycemic controls. Co-treatment with Cerebrolysin (40 mg/ml) attenuated this high glucose-induced effect in a concentration-dependent manner. In normoglycemic conditions, treatment with Cerebrolysin caused a dose-dependent increase in the mean AUC values. Cerebrolysin treatment resulted in maintenance of the functional integrity, survival, and promotion of neurite outgrowth of the cultured DRG neurons exposed to high glucose, indicating involvement of peripheral sensory neurons.

An Exploratory Study on the Contractile Effects of Hydroxy-α-Sanshool and Hydroxy-β-Sanshool, the Active Ingredients of Daikenchuto, on the Internal Anal Sphincter.

Daikenchuto (DKT) enhances the contraction of the internal anal sphincter (IAS) in patients with constipation and fecal incontinence; however, the mechanism of its action is unknown. We investigated the effects of the active ingredients of DKT (hydroxy-α-sanshool (HAS) and hydroxy-β-sanshool (HBS)) on the contractile activity of the canine rectum and IAS. Three male beagle dogs were prepared for each of the HAS, HBS, and control groups. Force transducers were attached to the rectal and IAS surfaces of the dogs, and the contractile responses were measured by telemetry under conscious conditions. HAS (10 mg/body) and HBS (2.5 mg/body) were administered intrarectally at doses previously identified from an effective dose of DKT extract (1.5 g/body), and contractile responses were recorded up to 6 h after administration. Contractile activity of the rectum and IAS was evaluated by observing the area under the curve (AUC) of the recorded contraction waveform. Plasma concentrations of HAS and HBS were measured before and after administration to confirm IAS exposure to both ingredients. The mean AUC values of the IAS for the control, HAS, and HBS groups at 10 min after administration were 115, 87, and 220 (g-min), respectively, indicating a higher contraction in the HBS group, which was maintained for approximately 3 h. As for the rectum, no contractile response was observed in either the HAS or HBS groups. Plasma concentrations of both ingredients peaked at 20 min after administration. HBS could be involved in the contractile action of DKT on the IAS.

Development of a Multivariate Prognostic Model for Lenvatinib Treatment in Hepatocellular Carcinoma.

Lenvatinib is a first-line agent for advanced hepatocellular carcinoma (HCC), but individual responses to treatment are highly heterogeneous. The aim of this study was to investigate the clinical parameters that influence the efficacy of Lenvatinib and to develop a prognostic model. We retrospectively enrolled 333 Lenvatinib-treated patients with HCC with a median age of 57 years. Two hundred nd sixty-three of these patients had BCLC (2022) stage C. The median overall survival (mOS) time within the cohort was 12.1 months, and the median progression-free survival (mPFS) time was 4.7 months. Univariate Cox regression, best subset regression, and Lasso regression were used to screen primary variables for possible contribution to OS, multivariate Cox analysis was used to fit selected models, and the final model was selected using the maximum area under the curve (AUC) and minimum AIC. Receiver operating curves (ROC), calibration curves, and decision curve analysis were plotted to assess model performance, and 5-fold cross-validation was performed for internal validation. X-tile software was used to select the best cutoff points and to divide the study cohort into 3 different risk groups. Seven variables were included in the final model: BCLC stage, prior transarterial chemoembolization and immunotherapy history, tumor number, prognostic nutritional index, log (alpha-fetoprotein), and log (platelet-to-lymphocyte ratio). We named this final model the "multivariate prognostic model for Lenvatinib" (MPML), and a nomogram was constructed to predict the probability of survival at 6, 9, and 12 months. The MPML had good discrimination, calibration, and applicability. Cross-validation showed mean AUC values of 0.7779, 0.7738, and 0.7871 at 6, 9, and 12 months, respectively. According to nomogram points, mOS time was 21.57, 8.70, and 5.37 months in the low, medium, and high-risk groups, respectively (P < .001), and these differences were also observed in the PFS survival curve (P < .001). The MPML stratified patients according to baseline clinical characteristics had a strong performance in predicting Lenvatinib efficacy and has the potential for use as an auxiliary clinical tool for individualized decision-making.

Abstract Number ‐ 51: Automated Versus Human Hyperdense Vessel Sign Detection Using Non‐Contrast Computed Tomography Scans

Introduction Rapid detection of large vessel occlusion (LVO) is very crucial in triaging stroke patients potentially candidates for mechanical thrombectomy (MT). Hyperdense vessel sign (HDVS) is one of the earliest ischemic changes in non‐contrast CT scan (NCCT) indicating LVO stroke. Artificial intelligence emerged to detect HDVS with the advantages of faster acquisition, less variation, and a lower need for experience than the usual detection. We aimed to identify the diagnostic performance of automated software (e‐Stroke, Brainomix) in HDVS detection. Methods A prospectively collectedMT database from March 2020 to August 2021 was reviewed. Patients were included if they had intracranial internal carotid artery or middle cerebral artery M1 or M2 occlusion. Cases with HDVS were identified through the routine 2.5‐mm slice thickness NCCT scans after being correlated with patients’ clinical information and confirmed with CT angiography (CTA) scans. NCCT scans were classified according to slice thickness into two groups: 2.5‐mm scans and 0.625‐mm generated scans. All NCCT scans were read by e‐Stroke software, then deidentified and reviewed by two stroke neurologists who were blinded to any clinical, other imaging, or therapeutic information. They were required to record the presence/laterality of HDVS before and after observing other NCCT early ischemic changes like gaze deviation, loss of insular ribbon, caudate or lentiform hypodensity. ROC curve analysis was used to estimate sensitivity and specificity and the area under the curve (AUC) was compared using DeLong’s test. Inter‐rater agreement between the two readers’ final reads, e‐Stroke, and the standard read was measured using the Fleiss Kappa test. Results Among 304 patients included in the study, 37.7% had HDVS. Approximately 44% of the scans had 2.5‐mm slice thickness and 56% had 0.625‐mm slice thickness. The e‐Stroke software identified HDVS with a sensitivity of 63% and a specificity of 71% (Table 1). The mean AUC value of e‐Stroke HDVS detection (0.67[0.61‐0.74]) was similar to reader‐1 (0.68[0.62‐0.74];p = 0.87) and reader‐2 (0.63[0.57‐0.70];p = 0.56). HDVS detection improved by reader‐1(0.78[0.72‐0.83];p = 0.03) after observing other early ischemic changes on the same scans, but reader‐2 performance remained similar to e‐Stroke (0.69[0.63‐0.76];p = 0.71). AUC, sensitivity and specificity ofHDVS detection by e‐Stroke were significantly higher using 2.5‐mm compared to 0.625‐mm sliced NCCT scans (0.78[0.70‐0.86],sensitivity 70%,specificity 86%;p&lt; 0.001) vs (0.58[0.50‐0.67],sensitivity 56%,specificity 61%;p = 0.06) respectively;p = 0.01. The readers also had higher AUC values with 2.5‐mm scans but not statistically significant, (0.74[0.66‐0.83] vs 0.64[0.56‐0.73];p = 0.18) for reader‐1 and (0.68[0.59‐0.77] vs 0.57[0.48‐0.66];p = 0.23) for reader‐2. The same after the final read, (0.85[0.78‐0.92] vs 0.75[0.67‐0.82];p = 0.08) for reader‐1 and (0.73[0.65‐0.82] vs 0.67[0.58‐0.76];p = 0.43) for reader‐2. Similarly, inter‐rater agreement was higher using 2.5‐mm sliced scans, k = 0.50(0.43‐0.75) compared to0.625‐mm scans,k = 0.27(0.21‐0.33). Conclusions Artificial intelligence (e‐Stroke software) has comparable sensitivity and specificity to human readers in HDVS detection. For e‐Stroke software, 2.5‐mm sliced CT scans are better to identifyHDVS compared to 0.625‐mm scans.

Radiation pneumonitis prediction model with integrating multiple dose-function features on 4DCT ventilation images

PurposeRadiation pneumonitis (RP) is dose-limiting toxicity for non-small-cell cancer (NSCLC). This study developed an RP prediction model by integrating dose-function features from computed four-dimensional computed tomography (4DCT) ventilation using the least absolute shrinkage and selection operator (LASSO). MethodsBetween 2013 and 2020, 126 NSCLC patients were included in this study who underwent a 4DCT scan to calculate ventilation images. We computed two sets of candidate dose-function features from (1) the percentage volume receiving > 20 Gy or the mean dose on the functioning zones determined with the lower cutoff percentile ventilation value, (2) the functioning zones determined with lower and upper cutoff percentile ventilation value using 4DCT ventilation images. An RP prediction model was developed by LASSO while simultaneously determining the regression coefficient and feature selection through fivefold cross-validation. ResultsWe found 39.3 % of our patients had a ≥ grade 2 RP. The mean area under the curve (AUC) values for the developed models using clinical, dose-volume, and dose-function features with a lower cutoff were 0.791, and the mean AUC values with lower and upper cutoffs were 0.814. The relative regression coefficient (RRC) on dose-function features with upper and lower cutoffs revealed a relative impact of dose to each functioning zone to RP. RRCs were 0.52 for the mean dose on the functioning zone, with top 20 % of all functioning zone was two times greater than that of 0.19 for these with 60 %–80 % and 0.17 with 40 %–60 % (P < 0.01). ConclusionsThe introduction of dose-function features computed from functioning zones with lower and upper cutoffs in a machine learning framework can improve RP prediction. The RRC given by LASSO using dose-function features allows for the quantification of the RP impact of dose on each functioning zones and having the potential to support treatment planning on functional image-guided radiotherapy.

Scoring Model for Predicting the Occurrence of Severe Illness in Hospitalized Patients with Severe Fever with Thrombocytopenia Syndrome.

Severe fever with thrombocytopenia syndrome (SFTS) is an emerging hemorrhagic fever with high mortality. Severe cases progressed rapidly, with deaths occurring within 2 weeks. Therefore, constructing a model to predict disease progression among hospitalized patients plays an important role in clinical practice. The development cohort included 121 patients with SFTS, 25 with severe SFTS, and 96 with mild SFTS. Two of the 64 variables were independent risk factors, including neurological symptoms (odds ratio [OR], 12.915; 95% confidence interval [CI], 3.342-49.916; P < 0.001) and aspartate aminotransferase/alanine aminotransferase levels (OR, 1.891; 95% CI, 1.272-2.813; P = 0.002). The model's area under the curve (AUC) was 0.882 (95% CI: 0.808-0.956). The mean AUC value obtained from the internal validation was 0.883 (95% CI: 0.809-0.957). The AUC in the external validation cohort was 0.873 (95% CI: 0.775-0.972). This model can be used to identify severely ill patients as early as possible with high predictive value, stability, and repeatability. This model can help clinicians with their treatment plans.

Quantifying the temporal stability in seasonal habitat for sage‐grouse using regression and ensemble tree approaches

AbstractIdentifying and quantifying the extent to which landscape‐level habitat variables drive the spatial distribution of individuals across a region can provide fundamental insights into a species ecology and be essential to wildlife management and conservation plans. Although the preferences for habitat resources and the resources themselves are not static over time, most research at large spatial scales does not consider seasonal effects nor quantify annual temporal variability in the spatial distribution of habitat resources. In this study, we used a machine learning (boosted regression trees [BRTs]) and generalized linear mixed model (GLMM) approach to quantify seasonal habitat selection across three life stages (nest, late brood, and winter habitat) of sage‐grouse and estimated annual stability across a 13‐year dataset in south‐central Wyoming. Generalized linear mixed models had high area under the curve (AUC) values, but were not as high as the BRT models that had mean AUC values of 0.86, 0.81, and 0.87 for nest, late brood, and winter habitat, respectively. Generalized linear mixed models and BRT result provided similar results, but because of the higher validation values of the BRT models, we assessed annual variation by predicting the BRT models across years. We found significant spatial trends in the distribution of nesting habitat, with general decreases in the relative probability of use across the core of the study area and corresponding increases in selection on the periphery. The primary temporally shifting variables for the nesting BRT models were development, Normalized Difference Vegetation Index, and topographic wetness, suggesting they were shifting out of preferable ranges for these variables as habitat suitability was decreased over the course of our study. Winter habitat appeared to have similar spatial changes in probability of selection, but these changes were likely related to changes in winter precipitation and snow depth, which were the primary contributors to the winter BRT models. The annual dynamics of habitat selection are seldom addressed in large‐scale research but can have potentially dramatic influences on our identification of preferred habitats.

Where do nivicolous myxomycetes occur? – Modeling the potential worldwide distribution of Physarum albescens

To identify potentially suitable areas for the mostly alpine ecological guild of nivicolous (snowbank) myxomycetes, the worldwide distribution of a distinct morphospecies, Physarum albescens, was modelled with a correlative spatial approach using the software MaxEnt from 537 unique occurrence points. Three models were developed, first with only the 19 bioclimatic variables plus elevation from the WorldClim database, second with regularization to correct for pseudo-absence, and third with additional categorical environmental layer on snow cover. All three models showed high mean AUC (area under the curve) values (>0.970). Output maps were comparable, with the third model perhaps the most realistic. For this model, snow cover, precipitation of the coldest quarter (of the year), and elevation predicted best the distribution of Ph. albescens. Elevation alone is a good predictor only in some regions, since (i) elevation of the occurrence points decreases with increasing latitude, and (ii) elevation wrongly predicts the species’ occurrence in arid mountain ranges. The model showed mountains in humid climates with highest incidence, which confirmed field studies: a long-lasting snow cover fluctuating with comparatively mild summers is the decisive factor. As such, the model can serve as a predictive map where fructifications of nivicolous myxomycetes can be expected. Limitations of the model are discussed: cryptic speciation within a morphospecies, including the evolution of reproductively isolated units which may lead to local adaptation and niche differentiation, and wider ranges for myxamoebal populations.

Predicting the Likelihood of Live Birth in Assisted Reproductive Technology According to the Number of Oocytes Retrieved and Female Age Using a Generalized Additive Model: A Retrospective Cohort Analysis of 17,948 Cycles.

CapsuleWe designed a predictive reference model to evaluate how many stimulation cycles are needed for a patient to achieve an ideal live birth rate using assisted reproductive technology.ObjectiveTo develop a counseling tool for women who wish to undergo assisted reproductive technology (ART) treatment to predict the likelihood of live birth based on age and number of oocytes retrieved.MethodsThis was a 6-year population-based retrospective cohort analysis using individual patient ART data. Between 2012 and 2017, 17,948 women were analyzed from their single ovarian stimulation cycle until they had a live birth or had used all their embryos. All consecutive women between 20 and 49 years old undergoing their ovarian stimulation cycles for ART in our center were enrolled. The cumulative live birth rate (CLBR) was defined as the delivery of a live neonate born during fresh or subsequent frozen–thawed embryo transfer cycles. Only the first delivery was considered in the analysis. Binary logistic regression was performed to identify and adjust for factors known to affect the CLBR independently. A generalized additive model was used to build a predictive model of CLBR according to the woman’s age and the number of oocytes retrieved.ResultsAn evidenced-based counseling tool was created to predict the probability of an individual woman having a live birth, based on her age and the number of oocytes retrieved in ART cycles. The model was verified by 10 times 10-fold cross-validation using the preprocessed data, and 100 area under the curve (AUC) values for receiver operating characteristic (ROC) curves were obtained on the test set. The mean AUC value was 0.7394. Our model predicts different CLBRs ranging from nearly 90% to less than 20% for women aged 20–49 years with at least 22 oocytes retrieved. The CLBRs of women aged 20–28 years were very similar, nearly on one trend line with a certain number of oocytes retrieved. Differences in the CLBR began to appear by the age of 29 years; these increased gradually in women aged >35 years.ConclusionA predictive model of the CLBR was designed to serve as a guide for physicians and for patients considering ART treatment. The number of oocytes needed to be retrieved to achieve a live birth depends on the woman’s age.

Deep neural networks with promising diagnostic accuracy for the classification of atypical femoral fractures

Background and purpose — A correct diagnosis is essential for the appropriate treatment of patients with atypical femoral fractures (AFFs). The diagnostic accuracy of radiographs with standard radiology reports is very poor. We derived a diagnostic algorithm that uses deep neural networks to enable clinicians to discriminate AFFs from normal femur fractures (NFFs) on conventional radiographs. Patients and methods — We entered 433 radiographs from 149 patients with complete AFF and 549 radiographs from 224 patients with NFF into a convolutional neural network (CNN) that acts as a core classifier in an automated pathway and a manual intervention pathway (manual improvement of image orientation). We tested several deep neural network structures (i.e., VGG19, InceptionV3, and ResNet) to identify the network with the highest diagnostic accuracy for distinguishing AFF from NFF. We applied a transfer learning technique and used 5-fold cross-validation and class activation mapping to evaluate the diagnostic accuracy. Results — In the automated pathway, ResNet50 had the highest diagnostic accuracy, with a mean of 91% (SD 1.3), as compared with 83% (SD 1.6) for VGG19, and 89% (SD 2.5) for InceptionV3. The corresponding accuracy levels for the intervention pathway were 94% (SD 2.0), 92% (2.7), and 93% (3.7), respectively. With regards to sensitivity and specificity, ResNet outperformed the other networks with a mean AUC (area under the curve) value of 0.94 (SD 0.01) and surpassed the accuracy of clinical diagnostics. Interpretation — Artificial intelligence systems show excellent diagnostic accuracies for the rare fracture type of AFF in an experimental setting.

Detection of epileptic seizures from compressively sensed EEG signals for wireless body area networks

Wireless body area networks (WBANs) are gaining popularity for tele-monitoring of biomedical signals such as the electroencephalogram (EEG), with diagnosing and monitoring of epileptic seizures being one of the most important applications. Most seizure-detection algorithms cannot be applied directly to the compressed data in WBANs as they require full reconstruction of original EEG signals. In this study, we propose a novel feature for real-time automatic single-channel seizure detection which does not require complete reconstruction of original EEGs. The feature is based on iteratively applying the orthogonal matching pursuit (OMP) algorithm on the compressed EEG data and computing the rate by which the energies of partially reconstructed signals are increased. The feature, i.e. partial energy difference (PED), is then used for classifying seizure and non-seizure states. We also extend this method to the case for multichannel EEG. In multichannel case, the simultaneous OMP (SOMP) with a low number of iterations (1 and 15 iteration) is applied to the compressed data and the difference between the Frobenius norms of partially reconstructed signals is used as a multivariate feature for the classification of seizure and non-seizure states. The proposed features are used to detect seizure intervals in the EEG database provided by the University of Bonn and the CHB-MIT database. The results show that the proposed features can classify seizure epochs from non-seizures even for compression ratios as small as 0.05. The results also show that the proposed single-channel method achieves improvement of up to 4% for the area under the curve (AUC) with significantly less execution time compared to the benchmark matrix determinant (MD) classification approach. The results of applying the proposed multivariate feature to seizure and non-seizure segments of length 5 s from the CHB-MIT database show that it achieves mean AUC values of 0.941, 0.941, and 0.939 with mean execution times of 9.5, 10.7, and 13.1 s for compression ratios of 0.05, 0.1, and 0.2, respectively. Applying a threshold to the PED in a leave-one-out cross-validation (LOO-CV) scenario generates a sensitivity of 0.873,specificity of 0.710, and accuracy of 0.791 at the compression ratio of 0.05. The proposed single- and multichannel features have the potential for deployment in WBANs for the real-time tele-monitoring of epilepsy patients in healthcare applications.

Diagnosis of Occlusal Tooth Wear Using 3D Imaging of Optical Coherence Tomography Ex Vivo.

The aim of this study was to assess the utility of 3D imaging of optical coherence tomography (OCT) for the diagnosis of occlusal tooth wear ex vivo. Sixty-three extracted human molars with or without visible tooth wear were collected to take digital intraoral radiography and 3D OCT images. The degree of tooth wear was evaluated by 12 examiners and scored using 4-rank scale: 1—slight enamel wear; 2—distinct enamel wear; 3—tooth wear with slight dentin exposure; 4—tooth wear with distinct involvement of dentin. The degree of tooth wear was validated by the histological view of confocal laser scanning microscopy (CLSM). The sensitivity, specificity, and area under the curve (AUC) of receiver operating characteristic analysis were calculated. Diagnostic accuracy was compared with the agreement with CLSM observation using weighted kappa. The results were statistically analyzed at a significance level of α = 0.05. Three-dimensional OCT showed significantly higher sensitivity (p < 0.05) for all the diagnostic thresholds of enamel wear and dentin exposure than digital radiography (0.82, 0.85, and 0.79 vs. 0.56, 0.52, and 0.57, respectively). Three-dimensional OCT showed higher AUC and kappa coefficients than digital radiography (p < 0.05), where mean AUC and Kappa values were 0.95 and 0.76 for OCT and 0.92 and 0.47 for radiography, respectively. No significant difference of specificity was observed (p > 0.05). Three-dimensional OCT could visualize and estimate the degree of tooth wear and detect the dentin exposure at the tooth wear surface accurately and reproducibly. Consequently, a new guideline for tooth wear assessment can be proposed using OCT.

Optimal slice thickness of brain computed tomography using a hybrid iterative reconstruction algorithm for identifying hyperdense middle cerebral artery sign of acute ischemic stroke.

To determine the optimal slice thickness of brain non-contrast computed tomography using a hybrid iterative reconstruction algorithm to identify hyperdense middle cerebral artery sign in patients with acute ischemic stroke. We retrospectively enrolled 30 patients who had presented hyperdense middle cerebral artery sign and 30 patients who showed no acute ischemic change in acute magnetic resonance imaging. Reformatted axial images at an angle of the orbitomeatal line in slice thicknesses of 0.5, 1, 3, 5, and 7mm were generated. Optimal slice thickness for identifying hyperdense middle cerebral artery sign was evaluated by a receiver operating characteristics curve analysis and area under the curve (AUC). The mean AUC value of 0.5-mm slice (0.921; 95% confidence interval (95% CI), 0.868 to 0.975) was significantly higher than those of 3-mm (0.791; 95% CI, 0.686 to 0.895; p = 0.041), 5-mm (0.691; 95% CI, 0.583 to 0.799, p < 0.001), and 7-mm (0.695; 95% CI, 0.593 to 0.797, p < 0.001) slices, whereas it was equivalent to that of 1-mm slice (0.901; 95% CI, 0.837 to 0.965, p = 0.751). Thin slice thickness of ≤ 1mm has a better diagnostic performance for identifying hyperdense artery sign on brain non-contrast computed tomography with a hybrid iterative reconstruction algorithm in patients with acute ischemic stroke.

ABCG2 C421A polymorphisms affect exposure of the epidermal growth factor receptor inhibitor gefitinib.

ATP-binding castle protein G2 (ABCG2) is thought to inhibit the activities of certain gefitinib transporters, thereby affecting drug pharmacokinetics. The C421A polymorphism affects the function and expression of ABCG2 on the cell membrane. Previous studies have shown that proton-pump inhibitors (PPIs) inhibit gefitinib absorption, as well as the function of ABCG2. We evaluated the plasma concentrations of gefitinib in patients with and without the ABCG2 C421A polymorphism, who were or were not taking PPIs. In total, 61 patients with advanced epidermal-growth-factor-positive non-small-cell lung cancer were enrolled in this study. They were treated with gefitinib at a dose of 250mg per day. Plasma gefitinib concentration and ABCG2 C421A status were determined after 2weeks. The patients were divided into CC- and CA/AA genotype groups. We compared the trough and peak gefitinib levels and the area under the curve (AUC) values for 24-h gefitinib concentrations. We also compared these parameters among four groups distinguished according to the presence or absence of the polymorphism and PPI use. The mean trough gefitinib level and AUC value for 24-h gefitinib concentration were significantly lower in the CA/AA group compared to the CC group (mean trough level: 333.2 vs. 454.5ng/mL, respectively, P = 0.021; AUC: 9949.9 vs. 13,085.4ng・h/mL, respectively, P = 0.034). Among patients taking PPIs, the mean trough gefitinib level was significantly lower in the CA/AA group than the CC group (220.1 vs. 340.5ng/mL, respectively, P = 0.033). The CA/AA-type of ABCG2 C421A polymorphism may be associated with lower gefitinib plasma concentrations.

Deep Learning in Thyroid Ultrasonography to Predict Tumor Recurrence in Thyroid Cancers.

PurposeTo evaluate a deep learning model to predict recurrence of thyroid tumor using preoperative ultrasonography (US).Materials and MethodsWe included representative images from 229 US-based patients (male:female = 42:187; mean age, 49.6 years) who had been diagnosed with thyroid cancer on preoperative US and subsequently underwent thyroid surgery. After selecting each representative transverse or longitudinal US image, we created a data set from the resulting database of 898 images after augmentation. The Python 2.7.6 and Keras 2.1.5 framework for neural networks were used for deep learning with a convolutional neural network. We compared the clinical and histological features between patients with and without recurrence. The predictive performance of the deep learning model between groups was evaluated using receiver operating characteristic (ROC) analysis, and the area under the ROC curve served as a summary of the prognostic performance of the deep learning model to predict recurrent thyroid cancer.ResultsTumor recurrence was noted in 49 (21.4%) among the 229 patients. Tumor size and multifocality varied significantly between the groups with and without recurrence (p < 0.05). The overall mean area under the curve (AUC) value of the deep learning model for prediction of recurrent thyroid cancer was 0.9 ± 0.06. The mean AUC value was 0.87 ± 0.03 in macrocarcinoma and 0.79 ± 0.16 in microcarcinoma.ConclusionA deep learning model for analysis of US images of thyroid cancer showed the possibility of predicting recurrence of thyroid cancer.

Prediction of malignant glioma grades using contrast-enhanced T1-weighted and T2-weighted magnetic resonance images based on a radiomic analysis

We conducted a feasibility study to predict malignant glioma grades via radiomic analysis using contrast-enhanced T1-weighted magnetic resonance images (CE-T1WIs) and T2-weighted magnetic resonance images (T2WIs). We proposed a framework and applied it to CE-T1WIs and T2WIs (with tumor region data) acquired preoperatively from 157 patients with malignant glioma (grade III: 55, grade IV: 102) as the primary dataset and 67 patients with malignant glioma (grade III: 22, grade IV: 45) as the validation dataset. Radiomic features such as size/shape, intensity, histogram, and texture features were extracted from the tumor regions on the CE-T1WIs and T2WIs. The Wilcoxon–Mann–Whitney (WMW) test and least absolute shrinkage and selection operator logistic regression (LASSO-LR) were employed to select the radiomic features. Various machine learning (ML) algorithms were used to construct prediction models for the malignant glioma grades using the selected radiomic features. Leave-one-out cross-validation (LOOCV) was implemented to evaluate the performance of the prediction models in the primary dataset. The selected radiomic features for all folds in the LOOCV of the primary dataset were used to perform an independent validation. As evaluation indices, accuracies, sensitivities, specificities, and values for the area under receiver operating characteristic curve (or simply the area under the curve (AUC)) for all prediction models were calculated. The mean AUC value for all prediction models constructed by the ML algorithms in the LOOCV of the primary dataset was 0.902 ± 0.024 (95% CI (confidence interval), 0.873–0.932). In the independent validation, the mean AUC value for all prediction models was 0.747 ± 0.034 (95% CI, 0.705–0.790). The results of this study suggest that the malignant glioma grades could be sufficiently and easily predicted by preparing the CE-T1WIs, T2WIs, and tumor delineations for each patient. Our proposed framework may be an effective tool for preoperatively grading malignant gliomas.