Unraveling gender-specific structural brain differences in drug-resistant epilepsy using advanced deep learning techniques

Increased expression of calponin-3 in epileptic patients and experimental rats

PurposeIndividuals with drug-resistant epilepsy may benefit from epilepsy surgery. In nonlesional cases, where no epileptogenic lesion can be detected on structural magnetic resonance imaging, multimodal neuroimaging studies are required. Breath-hold-triggered BOLD fMRI (bh-fMRI) was developed to measure cerebrovascular reactivity in stroke or angiopathy and highlights regional network dysfunction by visualizing focal impaired flow increase after vasodilatory stimulus. This regional dysfunction may correlate with the epileptogenic zone. In this prospective single-center single-blind pilot study, we aimed to establish the feasibility and safety of bh-fMRI in individuals with drug-resistant non-lesional focal epilepsy undergoing presurgical evaluation.MethodsIn this prospective study, 10 consecutive individuals undergoing presurgical evaluation for drug-resistant focal epilepsy were recruited after case review at a multidisciplinary patient management conference. Electroclinical findings and results of other neuroimaging were used to establish the epileptogenic zone hypothesis. To calculate significant differences in cerebrovascular reactivity in comparison to the normal population, bh-fMRIs of 16 healthy volunteers were analyzed. The relative flow change of each volume of interest (VOI) of the atlas was then calculated compared to the flow change of the whole brain resulting in an atlas of normal cerebral reactivity. Consequently, the mean flow change of every VOI of each patient was tested against the healthy volunteers group. Areas with significant impairment of cerebrovascular reactivity had decreased flow change and were compared to the epileptogenic zone localization hypothesis in a single-blind design.ResultsAcquisition of bh-fMRI was feasible in 9/10 cases, with one patient excluded due to noncompliance with breathing maneuvers. No adverse events were observed, and breath-hold for intermittent hypercapnia was well tolerated. On blinded review, we observed full or partial concordance of the local network dysfunction seen on bh-fMRI with the electroclinical hypothesis in 6/9 cases, including cases with extratemporal lobe epilepsy and those with nonlocalizing 18F-fluorodeoxyglucose positron emission tomography (FDG-PET).ConclusionThis represents the first report of bh-fMRI in individuals with epilepsy undergoing presurgical evaluation. We found bh-fMRI to be feasible and safe, with a promising agreement to electroclinical findings. Thus, bh-fMRI may represent a potential modality in the presurgical evaluation of epilepsy. Further studies are needed to establish clinical utility.Supplementary InformationThe online version of this article (10.1007/s00062-023-01363-2) contains supplementary material, which is available to authorized users.

Automated image classification is an essential task of the computer vision field. The tagging of images into a set of predefined groups is referred to as image classification. The implementation of computer vision to automate image classification would be beneficial because manual image evaluation and identification can be time-consuming, particularly when there are many images of different classes. Deep learning approaches are proven to overperform existing machine learning techniques in many fields in recent years, and computer vision is one of the most notable examples. The very deep neural network (VDCNN) is a powerful deep learning model for image classification, and this paper examines it briefly using MNIST handwritten digit dataset. This dataset is used to prove the efficacy of very deep neural networks over other deep learning models. The proposed study aims to comprehend the very deep neural network architecture used to accomplish a handwritten digit recognition task. The feasibility of the proposed model is evaluated using mean accuracy, validation accuracy, and standard deviation. The study results of the very deep neural network model are compared to a convolutional neural network and convolutional neural network with batch normalization. According to the results of the comparison study, very deep neural networks achieve high accuracy of 99.1% for a handwritten dataset. The outcome of the proposed work is used to interpret how well a very deep neural network performs when compared to the other two models of deep neural networks. This proposed architecture may be used to automate the classification of handwritten digits dataset.

Achieving energy efficiency (EE) fairness among heterogeneous mobile devices will become a crucial issue in future wireless networks. This paper investigates a deep learning (DL) approach for improving EE fairness performance in interference channels (IFCs) where multiple transmitters simultaneously convey data to their corresponding receivers. To improve the EE fairness, we aim to maximize the minimum EE among multiple transmitter–receiver pairs by optimizing the transmit power levels. Due to fractional and max-min formulation, the problem is shown to be non-convex, and, thus, it is difficult to identify the optimal power control policy. Although the EE fairness maximization problem has been recently addressed by the successive convex approximation framework, it requires intensive computations for iterative optimizations and suffers from the sub-optimality incurred by the non-convexity. To tackle these issues, we propose a deep neural network (DNN) where the procedure of optimal solution calculation, which is unknown in general, is accurately approximated by well-designed DNNs. The target of the DNN is to yield an efficient power control solution for the EE fairness maximization problem by accepting the channel state information as an input feature. An unsupervised training algorithm is presented where the DNN learns an effective mapping from the channel to the EE maximizing power control strategy by itself. Numerical results demonstrate that the proposed DNN-based power control method performs better than a conventional optimization approach with much-reduced execution time. This work opens a new possibility of using DL as an alternative optimization tool for the EE maximizing design of the next-generation wireless networks.

Signature is an essential biometric modality used in daily life to authenticate one person on his claim. Due to the widespread usage of stamps in office work, banks, forensic documents, etc., it captures the great interest of the researchers for the past many years. With the advancement in the technology and internet, automatic signature verification considered with renewed interest by the researchers. In this paper, we present the state-of-the-art signature verification mechanism and a novel method to perform signature verification in the RGB images. We train our model using advanced deep learning technique (Deep Neural network (DNN)) to differentiate between genuine and forged classes of signatures. The proposed algorithm evaluated on 4NSigComp2010 and 4NSigComp2012 datasets with the same experimental setup. The results showed that the proposed algorithm outperforms the systems presented in the competitions with an EER of 12 and 11.75, respectively.

Machine learning has already been used as a resource for disease detection and health care as a complementary tool to help with various daily health challenges. The advancement of deep learning techniques and a large amount of data-enabled algorithms to outperform medical teams in certain imaging tasks, such as pneumonia detection, skin cancer classification, hemorrhage detection, and arrhythmia detection. Automated diagnostics, which are enabled by images extracted from patient examinations, allow for interesting experiments to be conducted. This research differs from the related studies that were investigated in the experiment. These works are capable of binary categorization into two categories. COVID-Net, for example, was able to identify a positive case of COVID-19 or a healthy person with 93.3% accuracy. Another example is CHeXNet, which has a 95% accuracy rate in detecting cases of pneumonia or a healthy state in a patient. Experiments revealed that the current study was more effective than the previous studies in detecting a greater number of categories and with a higher percentage of accuracy. The results obtained during the model's development were not only viable but also excellent, with an accuracy of nearly 96% when analyzing a chest X-ray with three possible diagnoses in the two experiments conducted.

RationaleEpilepsy presents significant sex-based disparities in prevalence and manifestation. Epidemiological studies reveal that epilepsy is more prevalent in males, with lesional types being more common, whereas idiopathic generalized epilepsies are more frequently observed in females. These differences stress the importance of considering sex-specific factors in epilepsy diagnosis, treatment, and mechanistic research using preclinical models. To elucidate potential molecular differences that could explain these disparities and inform personalized treatment strategies, we conducted a proteomic analysis of epileptic brain tissues from both an experimental mouse model of genetic epilepsy and humans with drug-resistant epilepsy (DRE).MethodsWe employed mass spectrometry-based proteomic analysis on brain tissues from DRE patients and thePtenknockout (KO) mouse model of genetic epilepsy with focal cortical dysplasia. Mouse samples included hippocampi from adult wild-type (WT) andPtenKO mice (4-5 per group and sex). Human samples included temporal cortex from 12 DRE adult patients (7 males, 5 females) and 5 non-epileptic (NE) controls (2 males, 3 females). Brain biopsies were collected with patients’ informed consent under approved IRB protocols (Indiana University Health Biorepository). Proteomic profiles were analyzed using principal component analysis (PCA) along with volcano plots to identify significant changes in protein expression. The enrichment analysis of differentially expressed proteins was conducted by Gene Ontology (GO) and Kyoto Encyclopedia of Gene and Genomes (KEGG) pathway.ResultsPCA revealed distinct clustering of brain proteomes between epilepsy and control cases in both human and mice, with 390 proteins showing significant differences in human and 437 proteins in mouse samples. These proteins are primarily associated with ion channels, synaptic processes, and neuronal energy regulation. In the mouse model, males have more pronounced proteomic changes than females, with enrichment in metabolic pathways and VEGF signaling pathway, indicating a more severe vascular permeability impairment in males. In human DRE cases, 118 proteins were significantly changed by comparing epileptic females to males. Pathway analysis revealed changes in metabolic pathways and the HIF-1 signaling pathway, indicating that altered neuronal activity and inflammation may lead to increased oxygen consumption.ConclusionThese findings highlight significant differences between epilepsy and control brain samples in both humans and mice. Sex-specific analysis revealed distinct pathway enrichments between females and males, with males exhibiting a broader range of alterations, suggesting more extensive proteomic alterations. This study offers valuable insights into potential underlying mechanisms of epilepsy and underscores the importance of considering sex as a key factor in epilepsy research and therapeutic development.

Damage quantification in truss structures by limited sensor-based surrogate model

<p>Predicting commodity prices, particularly food prices, is a significant concern for various stakeholders, especially in regions that are highly sensitive to commodity price volatility. Historically, many machine learning models like autoregressive integrated moving average (ARIMA) and support vector machine (SVM) have been suggested to overcome the forecasting task. These models struggle to capture the multifaceted and dynamic factors influencing these prices. Recently, deep learning approaches have demonstrated considerable promise in handling complex forecasting tasks. This paper presents a novel long short-term memory (LSTM) network-based model for commodity price forecasting. The model uses five essential commodities namely bread, meat, milk, oil, and petrol. The proposed model focuses on advanced feature engineering which involves moving averages, price volatility, and past prices. The results reveal that our model outperforms traditional methods as it achieves 0.14, 3.04%, and 98.2% for root mean square error (RMSE), mean absolute percentage error (MAPE), and R-squared (R<sup>2</sup>), respectively. In addition to the simplicity of the model, which consists of an LSTM single-cell architecture that reduced the training time to a few minutes instead of hours. This paper contributes to the economic literature on price prediction using advanced deep learning techniques as well as provides practical implications for managing commodity price instability globally.</p>

Perfil clínico y sociodemográfico de la epilepsia en adultos de un centro de referencia de Colombia

Purpose: Violent subjects were demonstrated to exhibit abnormal brain structures; however, the brain changes may be different between criminals committing affective (VA), predatory violence (VP), and non-violence (NV). Therefore, the purpose of this study was to compare the differences in brain structures and psychological characteristics between VA, VP, and NV offenders. Methods: Twenty male criminal subjects (7 VP; 6 VA; and 7 NV) offenders; and twenty age-matched male healthy non-criminals were enrolled in this study. All subjects received psychological assessments as well as magnetic resonance imaging scans of the brain. Analysis of variance (ANOVA) was performed to understand the differences among four groups with Bonferroni correction. The voxel-based morphometry and voxel-wise diffusion tensor imaging analyses were performed to compare the gray matter (GM) volume and white matter (WM) integrity between the groups. In significant regions, a Spearman correlation analysis was performed to understand the relationship between the brain changes and psychological scores. Results: The ANOVA analysis showed that AUDIT scores were significantly different among four groups, but no significant group difference was noted after Bonferroni correction. The imaging comparisons further demonstrated that the VP and NV offenders exhibited significant alterations of WM and GM tissues in the rectus and superior temporal gyrus, respectively. In addition, the VP offenders exhibited greater GM volumes than VA offenders in the right middle frontal gyrus, and NV offenders had greater GM volumes than VP offenders in the bilateral thalamus. Conclusion: We concluded that the VA, VP, and NV groups exhibited different degrees of alterations in GM and WM tissues in regions involved in emotion and cognition.

Epilepsy is a common neurological disorder that results in seizures in patients of all ages. The frequency of seizure episodes can be controlled by prescribing anti-seizure drugs. Drug-resistant epilepsy is a condition where the seizures are uncontrolled by strong medications. Such patients are at a high risk of getting seizures frequently and prone to injuries due to sudden falls. Many countries prohibit epileptic patients from driving as sudden seizure attacks can cause loss of lives and property. In the past decades immense work has been carried out in the to monitor the seizure activity in patients and alert caregivers to extend help in emergencies. The study proposes a smart health care Internet of things framework to provide immediate help to the epileptic patient during an episode while travelling in a self-driving car. In the proposed framework, the seizure alert from a wearable device of the patient is transmitted to the control application via cloud. The control application also receives data from the vision, ultrasonic, and radar sensors. The critical information of seizure alert and the sensor data commands the car to force stop. The seizure google map location of the car is sent to the patient’s caregiver as well as the registered hospital. Many applications are being developed to provide luxury and comfort to fully automated car drivers. Far from providing luxury to the driver Emergency Rapid Response to Epileptic Seizure aims to propose a solution that could save the life of an epileptic patient who is drug-resistant or prone to frequent attacks despite severe medications. The experimental results on synchonization of clouds show that the minimum time is 30 sec 30 ms and maximum time is 31 sec 63 ms. The experimental results prove that its recommended to alert the patient's caregiver directly from control application rather than alerting via cloud.

Objective. To identify predictive factors of favourable outcomes in children with drug resistant epilepsy (DRE) associated with structural brain changes, who received surgical treatment. Methods. We performed a retrospective analysis of surgical treatment in 239 children with drug resistant epilepsy treated at the Pediatric Neurosurgery Department of the A. L. Polenov Russian Neurosurgical Institute from 2004 to 2014. The age of the patients varied between 2 and 18 years. The follow-up time ranged from 1 year to 10 years (on average 4.3 years). Results. According to magnetic resonance imaging (MRI), 223 (93.3%) patients had different structural brain changes. Among the patients, 47 (21.1%) had cerebral cicatrix, gliosis, cysts and atrophic changes, 33 (14.8%) had malformations of the cortical development, 43 (19.3%) had benign brain tumors, 47 (21.1%) had vascular malformations, 21 (9.4%) had arachnoid cysts, 12 (5.4%) had hippocampal sclerosis, 11 (4.9%) had phacomatosis, 5 (2.2%) had Rasmussen’s encephalitis, and 4 (1.8%) had hypothalamic hamartoma. Positron emission tomography revealed the areas of gray matter hypometabolism in the brain, which correspond to the areas of a metabolic deficiency around the epileptic focus. Favorable long-term outcomes of epilepsy surgery (class I according to Engel scale) were registered in 67.6% of the cases. The most favorable results (class I according to Engel scale) were achieved after lesionectomy (81.9%), temporal resections (78.4%), and hemispherectomy (72.7%).Conclusions. The most important predictive factors of favorable outcomes after epilepsy surgery are as follows: detection of epileptogenic morphological changes of the brain, completeness of resection of an epileptic and epileptogenic areas, and absense of seizures in the early postoperative period.

Background: Epilepsy is a chronic nervous system disease characterized by fluctuations between moments of stability and instability within nerve cells, which presents in forms of epileptic seizures. Most cases of epilepsy are controlled by Anti-seizure medication (ASM), but up to 30% can be drug-resistant, posing a higher risk of psychological and neurological impairment in children. In such cases, surgery presents itself as a therapeutic possibility. Objective: To conduct a systematic review on current literature of surgical outcomes in epileptic pediatric patients, with focus on efficacy, safety and long-term follow-up. Methods: This is a systematic literature review based on articles indexed in the United States National Library of Medicine (PubMed) published between 2019 and 2024. Were used “epilepsy”, “surgery” and “pediatric” as a descriptor. Initially, 57 articles were found of which 15 were used after applying an inclusion and exclusion criteria. Results: From the articles analyzed, all being systematic review and/or meta-analysis, were mainly composed of long-term follow up of patients as long as 10 years post-operatively, the full-scale intelligent quotient (FSIQ) and neurocognitive academic abilities assessment in pre and post procedure and comparisons with non-surgical groups. Conclusion: The review demonstrated that surgical intervention shows significant efficiency in improving quality of life and reducing seizure frequency in pediatric drug resistant epilepsy, with high percentage remaining seizure-free in a long-term follow-up. Among the patients, seizure-freedom was even more strongly correlated with shorter epilepsy duration, highlighting the importance of considering surgery early in those patient’s treatment process. Moreover, it was documented that after the procedure, patients mostly surpassed or maintained their neurocognitive functions, with a minority experiencing slight neurologic deficit, most of which were transient. Furthermore, there was no observed decline in core academic skills and in the FSIQ evaluation, and individuals were able to keep their preoperative performances, underscoring the safety implications of the procedure. By consolidating the current literature, it illustrates valuable insights and establishes a solid foundation for the significance of early referral to surgery and a personalized treatment approach in pediatric patients with drug-resistant epilepsy.

Software Defined Networking (SDN) is developing as a new solution for the development and innovation of the Internet. SDN is expected to be the ideal future for the Internet, since it can provide a controllable, dynamic, and cost-effective network. The emergence of SDN provides a unique opportunity to achieve network security in a more efficient and flexible manner. However, SDN also has original structural vulnerabilities, which are the centralized controller, the control-data interface and the control-application interface. These vulnerabilities can be exploited by intruders to conduct several types of attacks. In this paper, we propose a deep learning (DL) approach for a network intrusion detection system (DeepIDS) in the SDN architecture. Our models are trained and tested with the NSL-KDD dataset and achieved an accuracy of 80.7% and 90% for a Fully Connected Deep Neural Network (DNN) and a Gated Recurrent Neural Network (GRU-RNN), respectively. Through experiments, we confirm that the DL approach has the potential for flow-based anomaly detection in the SDN environment. We also evaluate the performance of our system in terms of throughput, latency, and resource utilization. Our test results show that DeepIDS does not affect the performance of the OpenFlow controller and so is a feasible approach.