A ResNet-LSTM hybrid model for predicting epileptic seizures using a pretrained model with supervised contrastive learning.

In this paper, we propose a method for predicting epileptic seizures using a pre-trained model utilizing supervised contrastive learning and a hybrid model combining residual networks (ResNet) and long short-term memory (LSTM). The proposed training approach encompasses three key phases: pre-processing, pre-training as a pretext task, and training as a downstream task. In the pre-processing phase, the data is transformed into a spectrogram image using short time Fourier transform (STFT), which extracts both time and frequency information. This step compensates for the inherent complexity and irregularity of electroencephalography (EEG) data, which often hampers effective data analysis. During the pre-training phase, augmented data is generated from the original dataset using techniques such as band-stop filtering and temporal cutout. Subsequently, a ResNet model is pre-trained alongside a supervised contrastive loss model, learning the representation of the spectrogram image. In the training phase, a hybrid model is constructed by combining ResNet, initialized with weight values from the pre-trained model, and LSTM. This hybrid model extracts image features and time information to enhance prediction accuracy. The proposed method's effectiveness is validated using datasets from CHB-MIT and Seoul National University Hospital (SNUH). The method's generalization ability is confirmed through Leave-one-out cross-validation. From the experimental results measuring accuracy, sensitivity, and false positive rate (FPR), CHB-MIT was 91.90%, 89.64%, 0.058 and SNUH was 83.37%, 79.89%, and 0.131. The experimental results demonstrate that the proposed method outperforms the conventional methods.

Atypical occurrence of anti-Ma2-associated encephalitis after breast cancer surgery and COVID-19.

In this report, we present a rare case of anti-Ma2-associated encephalitis concurrent with coronavirus disease 2019 (COVID-19) following breast cancer surgery. The patient exhibited minimal clinical symptoms of COVID-19 infection but developed seizures and altered mental status after surgery, leading to diagnosis of a classic paraneoplastic syndrome. This case highlights the possibility of paraneoplastic neurological syndrome even after cancer surgery and the need for careful consideration of post-acute infection syndromes when neurological symptoms occur following an infection.

Change of iron content in brain regions after intravenous iron therapy in restless legs syndrome: quantitative susceptibility mapping study.

STUDY OBJECTIVES: The pathomechanism of restless legs syndrome (RLS) is related to brain iron deficiency and iron therapy is effective for RLS; however, the effect of iron therapy on human brain iron state has never been studied with magnetic resonance imaging. This study aimed to investigate the change of brain iron concentrations in patients with RLS after intravenous iron therapy using quantitative susceptibility mapping (QSM). METHODS: We enrolled 31 RLS patients and 20 healthy controls. All participants underwent initial baseline (t0) assessment using brain magnetic resonance imaging, serum iron status, and sleep questionnaires including international RLS Study Group rating scale (IRLS). RLS patients underwent follow-up tests at 6 and 24 weeks (t1 and t2) after receiving 1000 mg ferric carboxymaltose. Iron content of region-of-interest on QSM images was measured for 13 neural substrates using the fixed-shaped method. RESULTS: RLS symptoms evaluated using IRLS were significantly improved after iron treatment (t0: 29.7 ± 6.5, t1: 19.5 ± 8.5, t2: 21.3 ± 10.1; p < .001). There was no significant difference in susceptibility values between the controls and RLS patients at t0. In the caudate nucleus, putamen, and pulvinar thalamus of RLS patients, the QSM values differed significantly for three timepoints (p = .035, .048, and .032, respectively). The post-hoc analysis revealed that the QSM values increased at t1 in the caudate nucleus (66.8 ± 18.0 vs 76.4 ± 16.6, p = .037) and decreased from t1 to t2 in the putamen (69.4 ± 16.3 vs 62.5 ± 13.6, p = .025). Changes in the QSM values for the pulvinar and caudate nuclei at t1 were positively and negatively correlated with symptomatic improvement, respectively (r = 0.361 and -0.466, respectively). CONCLUSIONS: Intravenous iron treatment results in changes in brain iron content which correlate to reductions in RLS severity. This suggests a connection between symptom improvement and the associated specific brain regions constituting the sensorimotor network.

Quantitative measurement of motor activity during sleep in isolated REM sleep behavior disorder patients using actigraphy before and after treatment with clonazepam.

STUDY OBJECTIVES: We conducted a prospective study to quantify motor activity during sleep measured by actigraphy before and after 3 months of treatment with clonazepam in patients with video-polysomnography (vPSG) confirmed isolated rapid eye movement (REM) sleep behavior disorder (iRBD). METHODS: The motor activity amount (MAA) and the motor activity block (MAB) during sleep were obtained from actigraphy. Then, we compared quantitative actigraphic measures with the results of the REM sleep behavior disorder questionnaire for the previous 3-month period (RBDQ-3M) and of the Clinical Global Impression-Improvement scale (CGI-I), and analyzed correlations between baseline vPSG measures and actigraphic measures. RESULTS: Twenty-three iRBD patients were included in the study. After medication treatment, large activity MAA dropped in 39% of patients, and the number of MABs decreased in 30% of patients when applying 50% reduction criteria. 52% of patients showed more than 50% improvement in either one. On the other hand, 43% of patients answered "much or very much improved" on the CGI-I, and RBDQ-3M was reduced by more than half in 35% of patients. However, there was no significant association between the subjective and objective measures. Phasic submental muscle activity during REM sleep was highly correlated with small activity MAA (Spearman's rho = 0.78, p < .001) while proximal and axial movements during REM sleep correlated with large activity MAA (rho = 0.47, p = .030 for proximal movements, rho = 0.47, p = .032 for axial movements). CONCLUSIONS: Our findings imply that quantifying motor activity during sleep using actigraphy can objectively assess therapeutic response in drug trials in patients with iRBD.

Performance Enhancement in Federated Learning by Reducing Class Imbalance of Non-IID Data.

Due to the distributed data collection and learning in federated learnings, many clients conduct local training with non-independent and identically distributed (non-IID) datasets. Accordingly, the training from these datasets results in severe performance degradation. We propose an efficient algorithm for enhancing the performance of federated learning by overcoming the negative effects of non-IID datasets. First, the intra-client class imbalance is reduced by rendering the class distribution of clients close to Uniform distribution. Second, the clients to participate in federated learning are selected to make their integrated class distribution close to Uniform distribution for the purpose of mitigating the inter-client class imbalance, which represents the class distribution difference among clients. In addition, the amount of local training data for the selected clients is finely adjusted. Finally, in order to increase the efficiency of federated learning, the batch size and the learning rate of local training for the selected clients are dynamically controlled reflecting the effective size of the local dataset for each client. In the performance evaluation on CIFAR-10 and MNIST datasets, the proposed algorithm achieves 20% higher accuracy than existing federated learning algorithms. Moreover, in achieving this huge accuracy improvement, the proposed algorithm uses less computation and communication resources compared to existing algorithms in terms of the amount of data used and the number of clients joined in the training.

Emotional and Environmental Factors Aggravating Dream Enactment Behaviors in Patients with Isolated REM Sleep Behavior Disorder.

Objective: To identify emotional and environmental factors that aggravate dream enactment behaviors (DEBs) in isolated rapid eye movement (REM) sleep behavior disorder (iRBD). Methods: In this cross-sectional study, a total of 96 polysomnography-confirmed iRBD patients (mean age, 68.5 years; men, 68%) and their caregivers completed questionnaires regarding potential aggravating factors related to DEBs, including emotion/feelings (stress, anger, anxiety, depressive mood, fatigue, pain), food (alcohol, caffeine, overeating in the evening, fasting/hunger), activities and sleep patterns (strenuous exercise, sex before bed, conflict/fighting, sleep deprivation, oversleeping, sleeping away from home, watching TV before bed), weather/environmental factors (cloudy or rainy weather, heat, cold, noise) and medication (skipping medication, taking hypnotics). Results: The patients reported that stress (61%) was the most aggravating factor for DEBs, followed by anxiety (56%), anger (51%), fatigue (49%), and watching TV before bed (46%). Similarly, the caregivers reported that these factors were most relevant to the aggravation of DEBs in the patients, although some factors were ranked differently. In the subgroup analyses, aggravating factors for DEBs did not differ by RBD symptom severity. Interestingly, the proportion of patients experiencing DEB aggravation by stress, anxiety and depressive mood was significantly higher in women than in men. Furthermore, depressed patients reported that stress and cloudy or rainy weather made DEBs worse than nondepressed patients. Conclusion: Our results suggest that DEBs in iRBD patients may be mainly aggravated by emotional factors. These negative effects appeared to be more prominent in female and depressed patients.

White matter tract-specific microstructural disruption is associated with depressive symptoms in isolated RBD.

OBJECTIVE: White matter (WM) tract-specific changes may precede gray matter loss in isolated rapid eye movement sleep behavior disorder (iRBD). We aimed to evaluate tract-specific WM changes using tract-specific statistical analysis (TSSA) and their correlation with clinical variables in iRBD patients. METHODS: This was a cross-sectional single-center study of 50 polysomnography-confirmed iRBD patients and 20 age- and sex-matched controls. We used TSSA to identify tract-specific fractional anisotropy (FA) and mean diffusivity (MD) in fourteen major fiber tracts and analyzed between-group differences in these values. Correlations between FA or MD values and clinical variables, including RBD symptom severity, depression and cognition, were evaluated. RESULTS: Patients with iRBD showed lower FA in the right anterior thalamic radiation (ATR) and higher MD in the bilateral ATR and right inferior fronto-occipital fasciculus (IF-OF) than controls after adjusting for age, sex, and years of education. MD values in the IF-OF positively correlated with scores on the Korean version of the Rapid Eye Movement Sleep Behavior Disorder Questionnaire-Hong Kong (RBDQ-KR, p = 0.042) and the Korean version of the geriatric depression scale (GDS-K, p = 0.002) in iRBD patients. Only GDS-K scores independently correlated with IF-OF MD values after adjusting for RBDQ-KR scores (adjusted p = 0.026). CONCLUSION: This study suggests WM microstructural disruption in the bilateral ATR and right IF-OF in patients with iRBD and that alterations in the IF-OF may contribute to depressive symptoms.

Conformational change of organic cofactor PLP is essential for catalysis in PLP-dependent enzymes.

Pyridoxal 5'-phosphate (PLP)-dependent enzymes are ubiquitous, catalyzing various biochemical reactions of approximately 4% of all classified enzymatic activities. They transform amines and amino acids into important metabolites or signaling molecules and are important drug targets in many diseases. In the crystal structures of PLP-dependent enzymes, organic cofactor PLP showed diverse conformations depending on the catalytic step. The conformational change of PLP is essential in the catalytic mechanism. In the study, we review the sophisticated catalytic mechanism of PLP, especially in transaldimination reactions. Most drugs targeting PLP-dependent enzymes make a covalent bond to PLP with the transaldimination reaction. A detailed understanding of organic cofactor PLP will help develop a new drug against PLP-dependent enzymes. [BMB Reports 2022; 55(9): 439-446].

Association of Nucleus Basalis of Meynert Functional Connectivity and Cognition in Idiopathic Rapid-Eye-Movement Sleep Behavior Disorder.

BACKGROUND AND PURPOSE: Cognitive impairments are common in isolated rapid-eye-movement sleep behavior disorder (iRBD), in which the cholinergic system may play an important role. This study aimed to characterize the cortical cholinergic activity using resting-state functional connectivity (FC) of the nucleus basalis of Meynert (NBM) according to the cognitive status of iRBD patients. METHODS: In this cross-sectional study, 33 patients with polysomnography-confirmed iRBD and 20 controls underwent neuropsychological evaluations and resting-state functional magnetic resonance imaging. Thirteen of the iRBD patients had mild cognitive impairment (iRBD-MCI), and the others were age-matched patients with normal cognition (iRBD-NC). The seed-to-voxel NBM-cortical FC was compared among the patients with iRBD-MCI, patients with iRBD-NC, and controls. Correlations between average values of significant clusters and cognitive function scores were calculated in the patients with iRBD. RESULTS: There were group differences in the FC of the NBM with the left lateral occipital cortex and lingual gyrus (adjusted for age, sex, and education level). The strength of FC was lower in the iRBD-MCI group than in the iRBD-NC and control groups (each post-hoc p<0.001). The average NBM-lateral occipital cortex FC was positively correlated with the memory-domain score in iRBD patients. CONCLUSIONS: The results obtained in this study support that cortical cholinergic activity is impaired in iRBD patients with MCI. FC between NBM and posterior regions may play a central role in the cognitive function of these patients.

Expanding Indications for a Ketogenic Diet as an Adjuvant Therapy in Adult Refractory Status Epilepticus: an Exploratory Study Using Moderation Analysis.

Refractory status epilepticus (RSE) requires multimodal treatment approaches to achieve rapid seizure cessation and neuroprotection. A ketogenic diet (KD) has demonstrated efficacy as a nutritional therapeutic option for adult RSE. However, the group of adult RSE patients who would benefit from adopting a KD needs to be determined to appropriately select the patients indicated for a KD. Therefore, we conducted a nonrandomized retrospective cohort study to explore the therapeutic efficacy of a KD by investigating the moderation effect of a KD on the association between the clinical characteristics of RSE patients and their functional outcomes. This study investigated 140 RSE patients, including 32 patients treated with a KD; among these patients, 28 (81%) achieved seizure cessation. We found that KD moderated the reduction in the modified Rankin scale (mRS) score at discharge among patients who were older, had higher seizure severity scores, were under continuous intravenous anesthetic therapy (CIVAD), and had super-RSE. Age and seizure severity scores, but not CIVAD or super-RSE, were associated with a KD-moderated change in mRS score at 3 months. Thus, we consider that our study provides evidence of a neuroprotective effect of KD in the most severe RSE patients with very few remaining therapeutic options, but future randomized controlled trials in these subgroups of KD patients are necessary.

Encephalitis with status epilepticus and stroke as complications of non-severe COVID-19 in a young female patient: a case report.

BACKGROUND: Neurological manifestations of COVID-19 are thought to be associated with the disease severity of COVID-19 and poor clinical outcomes. Dysregulated immune responses are considered to be mediating such complications. Our case illustrates multiple critical neurological complications simultaneously developed in a patient with non-severe COVID-19 and successful recovery with a multifaceted therapeutic approach. The cerebrospinal fluid (CSF) interleukin-6 (IL-6) level was temporally correlated with the clinical severity of the status epilepticus in our patient, suggesting a causal relationship. CASE PRESENTATION: A previously healthy 20-year-old female patient presented with a first-onset seizure. Concomitant non-severe COVID-19 pneumonia was diagnosed. CSF study showed lymphocytic pleocytosis with elevated IL-6 levels in CSF. During hospitalization under the diagnosis of autoimmune encephalitis, status epilepticus developed, and the seizure frequency was temporally correlated with the CSF IL-6 level. Furthermore, a new embolic stroke developed without a significant cardioembolic source. Contrary to the exacerbated COVID-19-associated neurological complications, COVID-19 pneumonia was cleared entirely. After treatment with antiseizure medications, antithrombotics, antiviral agents, and immunotherapy, the patient was discharged with near-complete recovery. CONCLUSION: Active serological, and radiological evaluation can be helpful even in non-severe COVID-19, and multidimensional treatment strategies, including immunotherapy, can successfully reverse the neurological complication.

ES-RED (Early Seizure Recurrence in the Emergency Department) Calculator: A Triage Tool for Seizure Patients.

Seizure is a common neurological presentation in patients visiting the emergency department (ED) that requires time for evaluation and observation. Timely decision and disposition standards for seizure patients need to be established to prevent overcrowding in the ED and achieve patients' safety. Here, we conducted a retrospective cohort study to predict early seizure recurrence in the ED (ES-RED). We randomly assigned 688 patients to the derivation and validation cohorts (2:1 ratio). Prediction equations extracted routine clinical and laboratory information from EDs using logistic regression (Model 1) and machine learning (Model 2) methods. The prediction equations showed good predictive performance, the area under the receiver operating characteristics curve showing 0.808 in Model 1 [95% confidential interval (CI): 0.761-0.853] and 0.805 in Model 2 [95% CI: 0.747-0.857] in the derivation cohort. In the external validation, the models showed strong prediction performance of 0.739 [95% CI: 0.640-0.824] in Model 1 and 0.738 [95% CI: 0.645-0.819] in Model 2. Intriguingly, the lowest quartile group showed no ES-RED after 6 h. The ES-RED calculator, our proposed prediction equation, would provide strong evidence for safe and appropriate disposition of adult resolved seizure patients from EDs, reducing overcrowding and delays and improving patient safety.

Interaction Between Glucocorticoid Receptors and FKBP5 in Regulating Neurotransmission of the Hippocampus.

FK501 binding protein 51 (FKBP5) is a stress response prolyl isomerase that inhibits the translocation of the glucocorticoid receptor (GR) heterocomplex to the nucleus. Previous studies have shown that the expression levels of FKBP5 are positively correlated with psychiatric disorders, including depression and post-traumatic stress disorder. In rodents, FKBP5 deletion in the brain leads to be resilient to stress-induced depression. The hippocampus is known to be one of the primary locations mediating stress responses in the brain by providing negative feedback signals to the hypothalamus-pituitary-adrenal gland axis. Therefore, we aimed to investigate the role of FKBP5 and its interaction with GRs in the hippocampus. We observed that FKBP5 deletion in the hippocampus resulted in a minimal change in synaptic transmission. In the hippocampus, GR activation alters the release probability in inhibitory synapses as well as the postsynaptic contribution of glutamate receptors in excitatory synapses; however, no such alterations were induced in the absence of FKBP5. FKBP5 deficiency causes insensitivity to activated GRs in the hippocampus suggesting that FKBP5 mediates synaptic changes caused by GR activation. Our study provides electrophysiological evidence of stress resilience observed in FKBP5-deficient mice.

Alerting network alteration in isolated rapid eye movement sleep behavior disorder patients with mild cognitive impairment.

OBJECTIVE: Mild cognitive impairment (MCI) was found in 30-50% of the isolated REM sleep behavior disorder (iRBD) patients. Furthermore, it is known that patients with Parkinson's disease have attention network defects. Given that iRBD is known to be the prodromal disease of α-synucleinopathies, our aim was to investigate whether there are attention network dysfunctions in iRBD patients following the presence of MCI. METHODS: 14 healthy controls, 48 iRBD patients, 24 with MCI and 24 without MCI, were included in this study. Attention network task (ANT) was used to assess alerting, orienting, and executive control networks. Event-related potentials (ERPs) and behavioral performances were recorded during the ANT. Parietal N1 and P3 components were analyzed to find effects of the three attention networks. RESULTS: IRBD patients without MCI showed neuropsychological, behavioral, and ERP results similar to those of healthy controls. On the other hand, iRBD patients with MCI showed a general decline in cognitive domains with no alerting effect (controls, p = 0.043; iRBD-noMCI, p = 0.014; iRBD-MCI, p = 0.130) while preserving orienting and executive control effect. Furthermore, iRBD patients with MCI had impairments in executive function and verbal memory domains, compared to iRBD patients without MCI. CONCLUSIONS: Our findings indicate that when cognition is reduced to MCI levels in iRBD patients, the attention network, especially the alerting component, is impaired. The attention network and cognition, on the other hand, can be preserved in iRBD patients due to the compensatory mechanism.

Deep Reinforcement Learning for UAV Trajectory Design Considering Mobile Ground Users.

A network composed of unmanned aerial vehicles (UAVs), serving as base stations (UAV-BS network), is emerging as a promising component in next-generation communication systems. In the UAV-BS network, the optimal positioning of a UAV-BS is an essential requirement to establish line-of-sight (LoS) links for ground users. A novel deep Q-network (DQN)-based learning model enabling the optimal deployment of a UAV-BS is proposed. Moreover, without re-learning of the model and the acquisition of the path information of ground users, the proposed model presents the optimal UAV-BS trajectory while ground users move. Specifically, the proposed model optimizes the trajectory of a UAV-BS by maximizing the mean opinion score (MOS) for ground users who move to various paths. Furthermore, the proposed model is highly practical because, instead of the locations of individual mobile users, an average channel power gain is used as an input parameter. The accuracy of the proposed model is validated by comparing the results of the model with those of a mathematical optimization solver.

Campylobacter fetus subspecies venerealis meningitis associated with a companion dog in a young adult: a case report.

BACKGROUND: Campylobacter spp., common commensals in the gastrointestinal tract of animals, especially poultry, can cause acute gastrointestinal illness in humans through animal-to-human transmission. Although Campylobacter fetus, especially subspecies fetus, rarely leads to systemic infections such as bacteremia in immunocompromised patients, it is unclear whether Campylobacter fetus subspecies venerealis (Cfv) causes infectious diseases in humans. CASE PRESENTATION: A 28-year-old man with a history of chronic alcoholism visited the emergency department with weakness of the left extremities. The patient was clinically diagnosed with community-acquired bacterial meningitis. The organism from the blood culture was subsequently identified as Campylobacter fetus. On phylogenetic analysis, the 16S rRNA sequence showed 99.93% similarity with other Cfv 16S rRNA sequences. The patient had no exposure to identifiable sources except for close contact with a companion dog, which could have been a possible source of transmission. CONCLUSIONS: This case suggests that Cfv could lead to human systemic infections such as meningitis and that companion animals, in addition to well-known animal hosts, could be sources of transmission.

Usefulness of arterial spin labeling perfusion as an initial evaluation of status epilepticus.

This study aimed to evaluate the sensitivity and prognostic value of arterial spin labeling (ASL) in a large group of status epilepticus (SE) patients and compare them with those of other magnetic resonance (MR) sequences, including dynamic susceptibility contrast (DSC) perfusion imaging. We retrospectively collected data of patients with SE in a tertiary center between September 2016 and March 2020. MR images were visually assessed, and the sensitivity for the detection of SE and prognostication was compared among multi-delay ASL, DSC, fluid-attenuated inversion recovery (FLAIR), and diffusion-weighted imaging (DWI). We included 51 SE patients and 46 patients with self-limiting seizures for comparison. Relevant changes in ASL were observed in 90.2% (46/51) of SE patients, a percentage higher than those for DSC, FLAIR, and DWI. ASL was the most sensitive method for initial differentiation between SE and self-limiting seizures. The sensitivity of ASL for detecting refractory SE (89.5%) or estimating poor outcomes (100%) was higher than those of other MR protocols or electroencephalography and comparable to those of clinical prognostic scores, although the specificity of ASL was very low as 9.4% and 15.6%, respectively. ASL showed a better ability to detect SE and predict the prognosis than other MR sequences, therefore it can be valuable for the initial evaluation of patients with SE.

Apolipoprotein E ε4 is not associated with cognitive impairment in patients with idiopathic REM sleep behavior disorder.

We analyzed 136 patients (age, 67.5 ± 6.9 years; men, 59.6%) with idiopathic rapid eye movement sleep behavior disorder (iRBD). The results of the neuropsychological tests were not significantly different between APOE ε4 carriers and noncarriers, suggesting that the APOE ε4 allele was not associated with cognitive impairment in iRBD.