A Learnable and Explainable Wavelet Neural Network for EEG Artifacts Detection and Classification.

Electroencephalography (EEG) artifacts are very common in clinical diagnosis and can heavily impact diagnosis. Manual screening of artifact events is labor-intensive with little benefit. Therefore, exploring algorithms for automatic detection and classification of EEG artifacts can significantly assist clinical diagnosis. In this paper, we propose a learnable and explainable wavelet neural network (WaveNet) for EEG artifact detection and classification. The model is powered by the wavelet decomposition block based on invertible neural network, which can extract signal features without information loss, and a tree generator for building wavelet tree structure automatically. They provide the model with good feature extraction capabilities and explainability. To evaluate the model's performance more fairly, we introduce the base point level matching score (BASE) and the Event-Aligned Compensation Scoring (EACS) at the event level as two metrics for model performance evaluation. On the challenging Temple University EEG Artifact (TUAR) dataset, our model outperforms other baselines in terms of F1-score for both artifact detection and classification tasks. The case study also validates the model's ability to offer explainability for predictions based on frequency band energy, suggesting potential applications in clinical diagnosis.

Identification of Transcription Factors of Santalene Synthase Gene Promoters and SaSSY Cis-Elements through Yeast One-Hybrid Screening in Santalum album L.

The main components of sandalwood heartwood essential oil are terpenoids, approximately 80% of which are α-santalol and ß-santalol. In the synthesis of the main secondary metabolites of sandalwood heartwood, the key gene, santalene synthase (SaSSY), can produce α-santalene and ß-santalene by catalyzed (E, E)-FPP. Furthermore, santalene is catalyzed by the cytochrome monooxygenase SaCYP736A167 to form sandalwood essential oil, which then produces a fragrance. However, the upstream regulatory mechanism of the key gene santalene synthase remains unclear. In this study, SaSSY (Sal3G10690) promoter transcription factors and SaSSY cis-elements were screened. The results showed that the titer of the sandalwood cDNA library was 1.75 × 107 CFU/mL, 80% of the inserted fragments identified by PCR were over 750 bp in length, and the positivity rate of the library was greater than 90%. The promoter region of the SaSSY gene was shown to have the structural basis for potential regulatory factor binding. After sequencing and bioinformatics analysis, we successfully obtained 51 positive clones and identified four potential SaSSY transcriptional regulators. Sal6G03620 was annotated as the transcription factor MYB36-like, and Sal8G07920 was annotated as the small heat shock protein HSP20 in sandalwood. Sal1G00910 was annotated as a hypothetical protein of sandalwood. Sal4G10880 was annotated as a homeobox-leucine zipper protein (ATHB-15) in sandalwood. In this study, a cDNA library of sandalwood was successfully constructed using a yeast one-hybrid technique, and the transcription factors that might interact with SaSSY gene promoters were screened. This study provides a foundation for exploring the molecular regulatory mechanism involved in the formation of sandalwood heartwood.

Ketogenic diet therapy in children with epilepsy caused by SLC2A1 mutations: a single-center single-arm retrospective study.

BACKGROUND: This retrospective study assessed the efficacy and safety of ketogenic diet therapies in children with epilepsy caused by SLC2A1 genetic mutations and glucose transporter type 1 deficiency syndrome. METHODS: Pediatric patients with epilepsy symptoms admitted to our medical center between January 2017 and October 2021 were included if they presented with an SLC2A1 genetic mutation on whole-exome sequencing. We analyzed the patients' convulsions and treatment with antiepileptic drugs. The patients were followed up at different time periods after ketogenic diet therapies. RESULTS: Six patients with SLC2A1 mutations were included in this study. The patients had seizures of different types and frequencies, and they took antiepileptic drugs to relieve their symptoms. They were then treated with a ketogenic diet for at least four months. We analyzed epilepsy control rates at 1, 2, 3, 6, and 12 months after ketogenic diet treatment. All patients were seizure-free within a month of receiving the diet therapy. All patients were followed up for six months, three were followed up for 12 months after the treatment, and there was no recurrence of epilepsy during this period. After antiepileptic drug withdrawal, none of the patients experienced seizure relapse when receiving ketogenic diet treatment alone. No severe adverse events occurred during the therapy. CONCLUSIONS: Ketogenic diet therapy is very effective and safe for the treatment of epilepsy caused by SLC2A1 mutations. Therefore, patients with glucose transporter type 1 deficiency syndrome caused by SLC2A1 mutations should begin ketogenic diet treatment as soon as possible.

Sleep Disturbances in Chinese Children with Epilepsy: Associations with Behavioral Problems and Quality of Life.

Objective: To investigate the association between sleep disturbances and behavioral problems as well as quality of life (QOL) in Chinese children with epilepsy. Methods: Caregivers of 167 epileptic children aged 3 to 12 years completed the Child Sleep Habits Questionnaire (CSHQ), the Strengths and Difficulties Questionnaire (SDQ), and the Pediatric Quality of Life Inventory (PedsQL™, 4.0 Core). Results: The prevalence of sleep disturbances (CSHQ total score >41) in epileptic children was 73.7% [95% CI (66.9%.80.4%)]. Epileptic children with sleep disturbances demonstrated more behavioral problems and lower QOL compared to those without sleep disturbances. Sleep disturbances such as sleep anxiety and daytime sleepiness were associated with more behavioral problems and lower QOL (p <0.05). Linear regression analyses showed that higher disturbance in sleep duration domain were associated with more behavioral problems, while higher sleep disordered breathing domains was associated with lower QOL (p <0.05). The interaction between sleep disturbances and behavioral problems in predicting QOL was not significant. The sensitivity analysis using 48 as an alternative cutoff for CSHQ total score obtained consistent results. Conclusion: Sleep disturbances occur frequently among Chinese children with epilepsy, and are associated with more behavioral problems and lower QOL. The sleep disturbance-QOL association is unlikely contingent on behavioral problems. This study highlights the necessity of evaluating and treating sleep disturbances multidimensionally among children with epilepsy to promote their whole health and wellbeing.

SQNN: A Spike-wave index Quantification Neural Network with a pre-labeling algorithm for epileptiform activity identification and quantification in children.

OBJECTIVE: Electrical status epilepticus during slow sleep (ESES) is a phenomenon identified by strong activation of epileptiform activity in the electroencephalogram (EEG) during sleep. For children disturbed by ESES, spike-wave index (SWI) is defined to quantify the epileptiform activity in the EEG during sleep. Accurate SWI quantification is important for clinical diagnosis and prognosis. To quantify SWI automatically, a deep learning method is proposed in this paper. APPROACH: Firstly, a pre-labeling algorithm (PreLA) composed of the adaptive wavelet enhanced decomposition and a slow-wave discrimination rule is designed to efficiently label the EEG signal. It enables the collection of large-scale EEG dataset with fine-grained labels. Then, an SWI Quantification Neural Network (SQNN) is constructed to accurately classify each sample point as normal or abnormal and to identify the abnormal events. SWI can be calculated automatically based on the total duration of abnormalities and the length of the signal. MAIN RESULTS: Experiments on two datasets demonstrate that the PreLA is effective and robust for labeling the EEG data and the SQNN accurately and reliably quantifies SWI without using any thresholds. The average estimation error of SWI is 3.12%, indicating that our method is more accurate and robust than experts and previous related works. The processing speed of SQNN is 100 times faster than that of experts. SIGNIFICANCE: Deep learning provides a novel approach to automatic SWI quantification and PreLA provides an easy way to label the EEG data with ESES syndromes. The results of the experiments indicate that the proposed method has a high potential for clinical diagnosis and prognosis of epilepsy in children.

SQNN: a spike-wave index quantification neural network with a pre-labeling algorithm for epileptiform activity identification and quantification in children.

Objective.Electrical status epilepticus during slow sleep (ESES) is a phenomenon identified by strong activation of epileptiform activity in the electroencephalogram (EEG) during sleep. For children disturbed by ESES, spike-wave index (SWI) is defined to quantify the epileptiform activity in the EEG during sleep. Accurate SWI quantification is important for clinical diagnosis and prognosis. To quantify SWI automatically, a deep learning method is proposed in this paper.Approach.Firstly, a pre-labeling algorithm (PreLA) composed of the adaptive wavelet enhanced decomposition and a slow-wave discrimination rule is designed to efficiently label the EEG signal. It enables the collection of large-scale EEG dataset with fine-grained labels. Then, an SWI quantification neural network (SQNN) is constructed to accurately classify each sample point as normal or abnormal and to identify the abnormal events. SWI can be calculated automatically based on the total duration of abnormalities and the length of the signal.Main results.Experiments on two datasets demonstrate that the PreLA is effective and robust for labeling the EEG data and the SQNN accurately and reliably quantifies SWI without using any thresholds. The average estimation error of SWI is 3.12%, indicating that our method is more accurate and robust than experts and previous related works. The processing speed of SQNN is 100 times faster than that of experts.Significance.Deep learning provides a novel approach to automatic SWI quantification and PreLA provides an easy way to label the EEG data with ESES syndromes. The results of the experiments indicate that the proposed method has a high potential for clinical diagnosis and prognosis of epilepsy in children.

Genetic Diagnosis Spectrum and Multigenic Burden of Exome-Level Rare Variants in a Childhood Epilepsy Cohort.

Childhood epilepsy is a considerably heterogeneous neurological condition with a high worldwide incidence. Genetic diagnosis of childhood epilepsy provides the most accurate pathogenetic evidence; however, a large proportion of highly suspected cases remain undiagnosed. Accumulation of rare variants at the exome level as a multigenic burden contributing to childhood epilepsy should be further evaluated. In this retrospective analysis, exome-level sequencing was used to depict the mutation spectra of 294 childhood epilepsy patients from Shanghai Children's Medical Center, Department of Neurology. Furthermore, variant information from exome sequencing data was analyzed apart from monogenic diagnostic purposes to elucidate the possible multigenic burden of rare variants related to epilepsy pathogenesis. Exome sequencing reached a diagnostic rate of 30.61% and identified six genes not currently listed in the epilepsy-associated gene list. A multigenic burden study revealed a three-fold possibility that deleterious missense mutations in ion channel and synaptic genes in the undiagnosed cohort may contribute to the genetic risk of childhood epilepsy, whereas variants in the gene categories of cell growth, metabolic, and regulatory function showed no significant difference. Our study provides a comprehensive overview of the genetic diagnosis of a Chinese childhood epilepsy cohort and provides novel insights into the genetic background of these patients. Harmful missense mutations in genes related to ion channels and synapses are most likely to produce a multigenic burden in childhood epilepsy.

External Cd2+ and protons activate the hyperpolarization-gated K+ channel KAT1 at the voltage sensor.

The functionally diverse cyclic nucleotide binding domain (CNBD) superfamily of cation channels contains both depolarization-gated (e.g., metazoan EAG family K+ channels) and hyperpolarization-gated channels (e.g., metazoan HCN pacemaker cation channels and the plant K+ channel KAT1). In both types of CNBD channels, the S4 transmembrane helix of the voltage sensor domain (VSD) moves outward in response to depolarization. This movement opens depolarization-gated channels and closes hyperpolarization-gated channels. External divalent cations and protons prevent or slow movement of S4 by binding to a cluster of acidic charges on the S2 and S3 transmembrane domains of the VSD and therefore inhibit activation of EAG family channels. However, a similar divalent ion/proton binding pocket has not been described for hyperpolarization-gated CNBD family channels. We examined the effects of external Cd2+ and protons on Arabidopsisthaliana KAT1 expressed in Xenopus oocytes and found that these ions strongly potentiate voltage activation. Cd2+ at 300 µM depolarizes the V50 of KAT1 by 150 mV, while acidification from pH 7.0 to 4.0 depolarizes the V50 by 49 mV. Regulation of KAT1 by Cd2+ is state dependent and consistent with Cd2+ binding to an S4-down state of the VSD. Neutralization of a conserved acidic charge in the S2 helix in KAT1 (D95N) eliminates Cd2+ and pH sensitivity. Conversely, introduction of acidic residues into KAT1 at additional S2 and S3 cluster positions that are charged in EAG family channels (N99D and Q149E in KAT1) decreases Cd2+ sensitivity and increases proton potentiation. These results suggest that KAT1, and presumably other hyperpolarization-gated plant CNBD channels, can open from an S4-down VSD conformation homologous to the divalent/proton-inhibited conformation of EAG family K+ channels.

Neutrophil to lymphocyte rate and serum prealbumin maybe predictors for abnormal high blood pressure caused by adrenocorticotropic hormone therapy in children with epileptic spasms: two cases report.

Epileptic spasms are a catastrophic form of epilepsy. When epileptic spasms occur under 2-year-old, they may be also called "infantile spasms". Adrenocorticotropic hormone (ACTH) is recommended as first line intervention for the treatment of epileptic spasms without tuberous sclerosis complex. The chief risks of ACTH therapy are immunosuppression and hypertension. We reported rare cases of abnormal high blood pressure in two male epileptic spasms patients during ACTH therapy. Both patients' blood pressure reached a high blood pressure stage 2 on the 9th day and 10th day of ACTH treatment, respectively. The blood pressure returned to normal range after the drug dosage was reduced or stopped. The lower level of neutrophil%, neutrophil count, and a higher level of lymphocyte%, lymphocyte count and prealbumin than normal range were observed in both patients before ACTH therapy. The neutrophil to lymphocyte rate might be a predictor for high blood pressure among patients treated with ACTH. The rates of both patients were under 0.50 (0.42 for Case 1 and 0.17 for Case 2). We reported the documented cases in two Chinese pediatric patients who suffered from epileptic spasms treated with ACTH resulted in abnormal high blood pressure, which could be predicted by using neutrophil to lymphocyte rate. We also mentioned serum prealbumin might be another predictor. More clinical data is required to elucidate the relationship between serum prealbumin level and blood pressure.

Novel TSC1 and TSC2 gene mutations in Chinese patients with tuberous sclerosis complex.

OBJECTIVE: The study was designed to identify pathogenic TSC1 or TSC2 gene mutations and provide solid evidence for the diagnosis of tuberous sclerosis complex (TSC). METHODS: 11 unrelated Chinese patients with TSC were investigated in the present study. Characteristic skin lesions such as hypomelanotic macules and the central nervous system features such as the epilepsy, cortical tubers and subependymal nodules were the most common symptoms that were observed in the patients. All exons and exon-intron boundaries of the TSC1 and TSC2 gene of the patients were amplified by PCR. RESULTS: A total of 11 different TSC2 and one TSC1 mutations were identified in the present study, of which five TSC2 and 1 TSC1 gene mutations were novel. Among the 11 patients, 10 harbored TSC2 mutations, whereas only one patient had a TSC1 gene mutation. The identification of TSC1/TSC2 gene mutations confirmed the diagnosis of the 11 patients with TSC. CONCLUSIONS: Our study has expanded the spectrum of TSC1 and TSC2 gene mutations causing TSC. The identification of the TSC1/TSC2 gene mutations confirmed the diagnosis of the 11 patients with TSC.