An accurate seizure prediction system enables early warnings before seizure onset of epileptic patients. It is extremely important for drug-refractory patients. Conventional seizure prediction works usually rely on features extracted from Electroencephalography (EEG) recordings and classification algorithms such as regression or support vector machine (SVM) to locate the short time before seizure onset. However, such methods cannot achieve high-accuracy prediction due to information loss of the hand-crafted features and the limited classification ability of regression and SVM algorithms. We propose an end-to-end deep learning solution using a convolutional neural network (CNN) in this paper. One and two dimensional kernels are adopted in the early-and late-stage convolution and max-pooling layers, respectively. The proposed CNN model is evaluated on Kaggle intracranial and CHB-MIT scalp EEG datasets. Overall sensitivity, false prediction rate, and area under receiver operating characteristic curve reaches 93.5%, 0.063/h, 0.981 and 98.8%, 0.074/h, 0.988 on two datasets respectively. Comparison with state-of-the-art works indicates that the proposed model achieves exceeding prediction performance.
Background Alcoholic liver disease (ALD) is one of the leading causes of chronic liver disease. Recent studies have demonstrated the roles of long noncoding RNAs (lncRNAs) in the pathogenesis of several disease processes. However, the roles of lncRNAs in patients with ALD remain unexplored. Methods Global profiling for human lncRNAs from peripheral blood RNA was performed in a well characterized cohort of healthy controls (HC, n=4), excessive drinkers without liver diseases (ED, n=4), and those with alcoholic cirrhosis with different severities (AC, n=12). The expression of unique lncRNA signatures were validated in a separate cohort of HC (n=17), ED (n=19), AC (n=48), and human liver tissues with ALD (n=19). Results Detailed analysis of plasma lncRNAs in AC subjects with different severities compared to HC identified 244 commonly up-regulated lncRNAs and 181 commonly down-regulated lncRNAs. We further validated top 20 most differentially up- and down-regulated lncRNAs in ED and AC as compared to HC and also determined the expression of selected lncRNAs in human liver tissues with or without AC. Among those lncRNAs, AK128652 and AK054921 were two of the most abundantly expressed lncRNAs in normal human plasma and liver, and their levels were significantly elevated in AC. The prognostic significance of AK128652 and AK054921 was determined in 48 subjects with AC; who were prospectively followed for 520 days. The expression of AK128652 and AK054921 was inversely associated with survival in patients with AC. Conclusions LncRNAs AK054921 and AK128652 are potential biomarkers to predict the progression to ALD in those with excessive alcohol consumption and are predictors of survival with patients with alcoholic cirrhosis.
BackgroundChronic subdural hematoma (CSDH) is a common disease that is more prevalent in older people. Surgical intervention is a safe treatment of choice. However, the recurrence rate is relatively high and the outcome is not always satisfactory among surgically treated patients. It is believed that aberrant angiogenesis and intracapsular inflammation contribute to the development of CSDH. Atorvastatin is reported to promote angiogenesis and suppress inflammation. We have recently shown that atorvastatin is effective to non-surgically reduce and eliminate CSDH with minimal side effects. Here, we report a clinical research trial protocol that is designed to evaluate the therapeutic effects of atorvastatin on CSDH.Methods/DesignWe have designed a multi-center, randomized, placebo-controlled, double blind clinical trial for evaluating the efficacy of oral atorvastatin in reducing CSDH. We have so far recruited 96 patients with CT-confirmed or MRI-confirmed CSDHs from 16 medical centers in China. These patients were originally recruited for the Oriental Neurosurgical Evidence-based Study Team (ONET) study. After informed consent is provided, patients are randomized to receive either atorvastatin (oral 20 mg/night for 8 weeks) or placebo (dextrin for 8 weeks); and followed for 16 weeks after the treatment. The primary outcome is the change in hematoma volume at the end of 8-week treatment. Secondary outcomes include: changes in 1) the hematoma volume at the 4th, 12th, and 24th weeks; 2) Markwalder’s Grading Scale and Glasgow Coma Scale (MGS-GCS); 3) Glasgow Outcome Score (GOS) and 4) Activities of Daily Life – the Barthel Index scale (ADL-BI). Safety will be assessed during the study by monitoring adverse events, laboratory tests, electrocardiography (ECG), measurements of vital signs (temperature, pulse, and blood pressure) and body weight.DiscussionResults of this trial will provide critical information regarding whether atorvastatin is an effective and safe alternative to surgical treatment of CSDH.Trial registrationClinicalTrials.gov Identifier – NCT02024373The date of trial registration: 7 August 2013Electronic supplementary materialThe online version of this article (doi:10.1186/s13063-015-1045-y) contains supplementary material, which is available to authorized users.
A new biometric technique using hand-dorsa vein patterns attracted attention these years. In this technique, extracting vein structures is a key procedure. For conventional algorithm, it is necessary to use high-quality images, which demand high-priced collection devices. That is because vein pattern images usually have low contrast, high-quality images are proper for segment. In our research, an extracting method for low-quality image is presented. The proposed method makes using low-cost devices possible. The results show that we could extract the vein networks as successfully as using high-quality images. In this paper, the principle of vein imaging is discussed, a new method to acquire vein images, which could enhance the contrast, is proposed, and the algorithm of extracting the vein pattern from low quality images is put forward. Our innovations are designing a new method to obtain good contrast vein images and proposing a novel denoising algorithm.Fourth International Conference on Image and Graphics 0-7695-2929-1/07 $25.00
Nowadays, several deep learning methods are proposed to tackle the challenge of epileptic seizure prediction. However, these methods still cannot be implemented as part of implantable or efficient wearable devices due to their large hardware and corresponding high-power consumption. They usually require complex feature extraction process, large memory for storing high precision parameters and complex arithmetic computation, which greatly increases required hardware resources. Moreover, available yield poor prediction performance, because they adopt network architecture directly from image recognition applications fails to accurately consider the characteristics of EEG signals. We propose in this paper a hardware-friendly network called Binary Single-dimensional Convolutional Neural Network (BSDCNN) intended for epileptic seizure prediction. BSDCNN utilizes 1D convolutional kernels to improve prediction performance. All parameters are binarized to reduce the required computation and storage, except the first layer. Overall area under curve, sensitivity, and false prediction rate reaches 0.915, 89.26%, 0.117/h and 0.970, 94.69%, 0.095/h on American Epilepsy Society Seizure Prediction Challenge (AES) dataset and the CHB-MIT one respectively. The proposed architecture outperforms recent works while offering 7.2 and 25.5 times reductions on the size of parameter and computation, respectively.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.