Integrating multi-task and cost-sensitive learning for predicting mortality risk of chronic diseases in the elderly using real-world data.

BACKGROUND AND OBJECTIVE: Real-world data encompass population diversity, enabling insights into chronic disease mortality risk among the elderly. Deep learning excels on large datasets, offering promise for real-world data. However, current models focus on single diseases, neglecting comorbidities prevalent in patients. Moreover, mortality is infrequent compared to illness, causing extreme class imbalance that impedes reliable prediction. We aim to develop a deep learning framework that accurately forecasts mortality risk from real-world data by addressing comorbidities and class imbalance. METHODS: We integrated multi-task and cost-sensitive learning, developing an enhanced deep neural network architecture that extends multi-task learning to predict mortality risk across multiple chronic diseases. Each patient cohort with a chronic disease was assigned to a separate task, with shared lower-level parameters capturing inter-disease complexities through distinct top-level networks. Cost-sensitive functions were incorporated to ensure learning of positive class characteristics for each task and achieve accurate prediction of the risk of death from multiple chronic diseases. RESULTS: Our study covers 15 prevalent chronic diseases and is experimented with real-world data from 482,145 patients (including 9,516 deaths) in Shenzhen, China. The proposed model is compared with six models including three machine learning models: logistic regression, XGBoost, and CatBoost, and three state-of-the-art deep learning models: 1D-CNN, TabNet, and Saint. The experimental results show that, compared with the other compared algorithms, MTL-CSDNN has better prediction results on the test set (ACC=0.99, REC=0.99, PRAUC=0.97, MCC=0.98, G-means = 0.98). CONCLUSIONS: Our method provides valuable insights into leveraging real-world data for precise multi-disease mortality risk prediction, offering potential applications in optimizing chronic disease management, enhancing well-being, and reducing healthcare costs for the elderly population.

RNA quality control factors nucleate Clr4/SUV39H and trigger constitutive heterochromatin assembly.

In eukaryotes, the Suv39 family of proteins tri-methylate lysine 9 of histone H3 (H3K9me) to form constitutive heterochromatin. However, how Suv39 proteins are nucleated at heterochromatin is not fully described. In the fission yeast, current models posit that Argonaute1-associated small RNAs (sRNAs) nucleate the sole H3K9 methyltransferase, Clr4/SUV39H, to centromeres. Here, we show that in the absence of all sRNAs and H3K9me, the Mtl1 and Red1 core (MTREC)/PAXT complex nucleates Clr4/SUV39H at a heterochromatic long noncoding RNA (lncRNA) at which the two H3K9 deacetylases, Sir2 and Clr3, also accumulate by distinct mechanisms. Iterative cycles of H3K9 deacetylation and methylation spread Clr4/SUV39H from the nucleation center in an sRNA-independent manner, generating a basal H3K9me state. This is acted upon by the RNAi machinery to augment and amplify the Clr4/H3K9me signal at centromeres to establish heterochromatin. Overall, our data reveal that lncRNAs and RNA quality control factors can nucleate heterochromatin and function as epigenetic silencers in eukaryotes.

End-to-End Multi-task Learning Architecture for Brain Tumor Analysis with Uncertainty Estimation in MRI Images.

Brain tumors are a threat to life for every other human being, be it adults or children. Gliomas are one of the deadliest brain tumors with an extremely difficult diagnosis. The reason is their complex and heterogenous structure which gives rise to subjective as well as objective errors. Their manual segmentation is a laborious task due to their complex structure and irregular appearance. To cater to all these issues, a lot of research has been done and is going on to develop AI-based solutions that can help doctors and radiologists in the effective diagnosis of gliomas with the least subjective and objective errors, but an end-to-end system is still missing. An all-in-one framework has been proposed in this research. The developed end-to-end multi-task learning (MTL) architecture with a feature attention module can classify, segment, and predict the overall survival of gliomas by leveraging task relationships between similar tasks. Uncertainty estimation has also been incorporated into the framework to enhance the confidence level of healthcare practitioners. Extensive experimentation was performed by using combinations of MRI sequences. Brain tumor segmentation (BraTS) challenge datasets of 2019 and 2020 were used for experimental purposes. Results of the best model with four sequences show 95.1% accuracy for classification, 86.3% dice score for segmentation, and a mean absolute error (MAE) of 456.59 for survival prediction on the test data. It is evident from the results that deep learning-based MTL models have the potential to automate the whole brain tumor analysis process and give efficient results with least inference time without human intervention. Uncertainty quantification confirms the idea that more data can improve the generalization ability and in turn can produce more accurate results with less uncertainty. The proposed model has the potential to be utilized in a clinical setup for the initial screening of glioma patients.

Analysis of a Super-Complex at Contact Sites Between Mitochondria and Plastids.

Plastids are organelles playing fundamental roles in different cellular processes such as energy metabolism or lipid biosynthesis. To fulfill their biogenesis and their function in the cell, plastids have to communicate with other cellular compartments. This communication can be mediated by the establishment of direct contact sites between plastids envelop and other organelles. These contacts are dynamic structures regulated in response to stress. For example, during phosphate (Pi) starvation, the number of contact sites between plastids and mitochondria significantly increases. In this situation, these contacts play an important role in the transfer of galactoglycerolipids from plastids to mitochondria. Recently, Pi starvation stress was used to identify key proteins involved in the traffic of galactoglycerolipids from plastids to mitochondria in Arabidopsis thaliana. A mitochondrial lipoprotein complex called MTL (Mitochondrial Transmembrane Lipoprotein) was identified. This complex contains mitochondrial proteins but also proteins located in the plastid envelope, suggesting its presence at the plastid-mitochondria junction. This chapter describes the protocol to isolate the MTL complex by clear-native polyacrylamide gel electrophoresis (CN-PAGE) from the mitochondrial fraction of Arabidopsis cell cultures and the methods to study different features of this complex.

Genome-Wide Identification and Expression Analysis of the DMP and MTL Genes in Sweetpotato (Ipomoea batatas L.).

Sweetpotato (Ipomoea batatas L.) is a strategic crop with both economic and energy value. However, improving sweetpotato varieties through traditional breeding approaches can be a time-consuming and labor-intensive process due to the complex genetic nature of sweetpotato as a hexaploid species (2n = 6x = 90). Double haploid (DH) breeding, based on in vivo haploid induction, provides a new approach for rapid breeding of crops. The success of haploid induction can be achieved by manipulating specific genes. Two of the most critical genes, DMP (DUF679 membrane proteins) and MTL (MATRILINEAL), have been shown to induce haploid production in several species. Here, we identified and characterized DMP and MTL genes in sweetpotato using gene family analysis. In this study, we identified 5 IbDMPs and 25 IbpPLAs. IbDMP5 and IbPLAIIs (IbPLAIIκ, IbPLAIIλ, and IbPLAIIµ) were identified as potential haploid induction (HI) genes in sweetpotato. These results provide valuable information for the identification and potential function of HI genes in sweetpotato and provide ideas for the breeding of DH lines.

Joint condition monitoring framework of wind turbines based on multi-task learning with poor-quality data.

Effective condition monitoring can improve the reliability of the turbine and reduce its downtime. However, due to the complexity of the operating conditions, the monitoring data is always mixed with poor-quality data. Poor-quality data mixed in monitoring tasks disrupts long-term dependency on data, which challenges traditional condition monitoring methods to work. To solve it, a joint reparameterization feature pyramid network (JRFPN) is proposed. Firstly, three different reparameterization tricks are designed to reform temporal information and exchange cross-temporal information, to alleviate the damage of long-term dependency. Secondly, a joint condition monitoring framework is designed, aiming to suppress feature confounding between poor-quality data and faulty data. The auxiliary task is trained to extract the degradation trend. The main task fights against feature confounding and dynamically delineates the failure threshold. The degradation trend and failure threshold decisions are corrected for each other to make the final joint state inference. Besides, considering the different quality of the monitoring variables, a channel weighting mechanism is designed to strengthen the ability of JRFPN. The measured data proved that JRFPN is more effective than other methods.

Predicting cell-type specific disease genes of diabetes with the biological network.

Type 2 diabetes (T2D) is a chronic condition that can lead to significant harm, such as heart disease, kidney disease, nerve damage, and blindness. Although T2D-related genes have been identified through Genome-wide association studies (GWAS) and various computational methods, the biological mechanism of T2D at the cell type level remains unclear. Exploring cell type-specific genes related to T2D is essential to understand the cellular mechanisms underlying the disease. To address this issue, we introduce DiGCellNet (predicting Disease Genes with Cell type specificity based on biological Networks), a model that integrates graph convolutional network (GCN) and multi-task learning (MTL) to predict T2D-associated cell type-specific genes based on the biological network. Our work represents the first attempt to predict cell type-specific disease genes using GCN and MTL. We evaluate our approach by predicting genes specific to four cell types and demonstrate that the proposed DiGCellNet outperforms other models that combine node embeddings with traditional machine learning algorithms. Moreover, DiGCellNet successfully identifies CALM1 as a gene specific to beta cell type in T2D cases, and this association is confirmed using an independent dataset. The code is available at https://github.com/23AIBox/23AIBox-DiGCellNet.

Recalling gist memory depends on CA1 hippocampal neurons for lifetime retention and CA3 neurons for memory precision.

Why some of us remember events more clearly than others and why memory loses precision over time is a major focus in memory research. Here, we show that the recruitment of specific neuroanatomical pathways within the medial temporal lobe (MTL) of the brain defines the precision of the memory recalled over the lifespan. Using optogenetics, neuronal activity mapping, and studying recent to very remote memories, we report that the hippocampal subfield CA1 is necessary for retrieving the gist of events and receives maximal support from MTL cortical areas (MEC, LEC, PER, and POR) for recalling the most remote memories. In contrast, reduction of CA3's activity alone coincides with the loss of memory precision over time. We propose that a shift between specific MTL subnetworks over time might be a fundamental mechanism of memory consolidation.

No convincing evidence the hippocampus is associated with working memory.

In a previous discussion paper , twenty-six working memory fMRI studies that reported activity in the hippocampus were systematically analyzed. None of these studies provided convincing evidence that the hippocampus was active during the late delay phase, the only period in which working memory can be isolated from long-term memory processes. Based on these results, it was concluded that working memory does not activate the hippocampus. Six commentaries on the discussion paper were received from Courtney (2022), Kessels and Bergmann (2022), Peters and Reithler (2022), Rose and Chao (2022), Stern and Hasselmo (2022), and Wood et al. (2022). Based on these commentaries, the present response paper considered whether there is evidence of sustained hippocampal activity during the working memory delay period based on depth-electrode recording, whether there are activity-silent working memory mechanisms in the hippocampus, and whether there is hippocampal lesion evidence indicating this region is important for working memory. There was no convincing electrophysiological or neuropsychological evidence that the hippocampus is associated with working memory maintenance, and activity-silent mechanisms were arguably speculative. Given that only a small fraction (approximately 5%) of working memory fMRI studies have reported hippocampal activity and lesion evidence indicates the hippocampus is not necessary for working memory, the burden of proof is on proponents of the view that the hippocampus is important for working memory to provide compelling evidence to support their position. To date, from my perspective, there is no convincing evidence that the hippocampus is associated with working memory.

Multi-task learning for toxic comment classification and rationale extraction.

Social media content moderation is the standard practice as on today to promote healthy discussion forums. Toxic span prediction is helpful for explaining the toxic comment classification labels, thus is an important step towards building automated moderation systems. The relation between toxic comment classification and toxic span prediction makes joint learning objective meaningful. We propose a multi-task learning model using ToxicXLMR for bidirectional contextual embeddings of input text for toxic comment classification, and a Bi-LSTM CRF layer for toxic span or rationale identification. To enable multi-task learning in this domain, we have curated a dataset from Jigsaw and Toxic span prediction datasets. The proposed model outperformed the single task models on the curated and toxic span prediction datasets with 4% and 2% improvement for classification and rationale identification, respectively. We investigated the domain adaptation ability of the proposed MTL model on HASOC and OLID datasets that contain the out of domain text from Twitter and found a 3% improvement in the F1 score over single task models.

Clinical Efficacy of Huanglian Jiedutang in Adjuvant Treatment of Acute Cerebral Infarction Complicated with Gastric Motility Disorder / 中国实验方剂学杂志

ObjectiveTo explore the clinical efficacy of Huanglian Jiedutang as an adjunctive treatment for acute cerebral infarction complicated with gastric motility disorder. MethodSixty patients with acute cerebral infarction complicated with gastric motility disorder with fire toxin syndrome were randomly divided into a western medicine control group （control group） and a traditional Chinese medicine （TCM） combined treatment group （observation group）， with 30 cases in each group. The control group received basic treatment for cerebral infarction and relevant western medical symptomatic treatment based on the patients' gastrointestinal symptoms. The observation group received Huanglian Jiedutang in addition to the treatment provided to the control group. The treatment course was 7 days. Neurological deficit scores and gastrointestinal dysfunction scores were assessed in both groups before treatment and on the 4th and 7th days of treatment. Gastrointestinal electrographic parameters， serum citrulline （CIT）， and motilin （MTL） levels were measured in both groups before treatment and on the 7th day of treatment. Clinical efficacy was compared between the two groups. ResultCompared with the baseline in both groups， the neurological deficit scores and gastrointestinal dysfunction scores were significantly reduced on the 4th and 7th days of treatment （P<0.05）. The reductions in these scores were more significant on the 7th day compared with those on the 4th day of treatment （P<0.05）. On the 4th and 7th days of treatment， the observation group showed a significantly greater reduction in neurological deficit scores and gastrointestinal dysfunction scores compared with the control group （P<0.05）. On the 7th day of treatment， compared with the baseline， both groups showed a significant increase in gastric antral and gastric body electric wave amplitudes as well as serum CIT and MTL levels （P<0.05）， and there were no statistically significant differences in the frequency of gastric antral and gastric body electric waves. On the 7th day of treatment， compared with the control group， the observation group had a significant increase in gastric antral and gastric body electric wave amplitudes as well as serum CIT and MTL levels （P<0.05）， and there were no statistically significant differences in the frequency of gastric antral and gastric body electric waves. After 7 days of treatment， the total effective rate in the observation group was 90.00% （27/30）， higher than 76.67% （23/30） in the control group， but the difference was not statistically significant. ConclusionAdjunctive treatment with Huanglian Jiedutang can effectively improve the symptoms of neurological function impairment and gastrointestinal dysfunction in patients with acute cerebral infarction complicated with gastric motility disorder， increase gastric antral and gastric body electric wave amplitudes， improve gastric motility disorder， and increase serum CIT and MTL levels， thereby improving the imbalanced secretion function of the gastrointestinal tract.

Trade-off between conservation of biological variation and batch effect removal in deep generative modeling for single-cell transcriptomics.

BACKGROUND: Single-cell RNA sequencing (scRNA-seq) technology has contributed significantly to diverse research areas in biology, from cancer to development. Since scRNA-seq data is high-dimensional, a common strategy is to learn low-dimensional latent representations better to understand overall structure in the data. In this work, we build upon scVI, a powerful deep generative model which can learn biologically meaningful latent representations, but which has limited explicit control of batch effects. Rather than prioritizing batch effect removal over conservation of biological variation, or vice versa, our goal is to provide a bird's eye view of the trade-offs between these two conflicting objectives. Specifically, using the well established concept of Pareto front from economics and engineering, we seek to learn the entire trade-off curve between conservation of biological variation and removal of batch effects. RESULTS: A multi-objective optimisation technique known as Pareto multi-task learning (Pareto MTL) is used to obtain the Pareto front between conservation of biological variation and batch effect removal. Our results indicate Pareto MTL can obtain a better Pareto front than the naive scalarization approach typically encountered in the literature. In addition, we propose to measure batch effect by applying a neural-network based estimator called Mutual Information Neural Estimation (MINE) and show benefits over the more standard maximum mean discrepancy measure. CONCLUSION: The Pareto front between conservation of biological variation and batch effect removal is a valuable tool for researchers in computational biology. Our results demonstrate the efficacy of applying Pareto MTL to estimate the Pareto front in conjunction with applying MINE to measure the batch effect.

Hippocampal intracerebral evoked potentials as a marker of its functionality in drug-resistant epilepsy.

OBJECTIVES: To assess hippocampal function during stereoelectroencephalography (SEEG) investigations through the study of the medial temporal lobe event-related potential (ERP) MTL-P300. METHODS: We recorded the MTL-P300 during a visual oddball task, using hippocampal electrodes implanted for SEEG in 71 patients, in a preoperative epilepsy investigation. The presence of an MTL-P300 and its amplitude were correlated with hippocampal involvement during seizures and memory function. RESULTS: Analysis using ROC curves revealed that an MTL-P300 amplitude below -46 µV, has a specificity of 93.3% in detecting the epileptogenic zone, and absence of the MTL-P300 in the left hippocampus of patients with typical language organization was associated with marked alteration of verbal memory scores. There was a significant correlation between performance in non-verbal memory tests and the amplitude of the MTL-P300 in the right hippocampus of patients with left hemispheric seizures (immediate visual recall: r = 0.67, p = 0.005; delayed visual recall: r = 0.56, p = 0.025). Using a linear regression, we confirmed that the absence of the MTL-P300 in the left hippocampus, the involvement of the left hippocampus during seizures, and the duration of epilepsy were predictors of verbal memory deficits. CONCLUSION: Analysis of the MTL-P300 during SEEG recording provides relevant information for the analysis of hippocampal functionality and can help to localize the epileptogenic zone.

Does working memory activate the hippocampus during the late delay period?

The aim of the present discussion paper was to identify whether any fMRI studies have provided convincing evidence that the hippocampus is associated with working memory. The key outcome variable was the phase in which hippocampal activity was observed: study, early delay, late delay, and/or test. During working memory tasks, long-term memory processes can operate during the study phase, early delay phase (due to extended encoding), or test phase. Thus, working memory processes can be isolated from long-term memory processes during only the late delay period. Twenty-six working memory studies that reported hippocampal activity were systematically analyzed. Many experimental protocol and analysis parameters were considered including number of participants, stimulus type(s), number of items during the study phase, delay duration, task during the test phase, behavioral accuracy, relevant fMRI contrast(s), whether the information was novel or familiar, number of phases modeled, and whether activation timecourses were extracted. For studies that were able to identify activity in different phases, familiar information sometimes produced activity during the study phase and/or test phase, but never produced activity during the delay period. When early-delay phase and late-delay phase activity could be distinguished via modeling these phases separately or inspecting activation timecourses, novel information could additionally produce activity during the early delay phase. There was no convincing evidence of hippocampal activity during the late delay period. These results indicate that working memory does not activate the hippocampus and suggest a model of working memory where maintenance of novel information can foster long-term memory encoding.

Differences in morbidity and mortality between unilateral adrenalectomy for adrenal Cushing's syndrome and bilateral adrenalectomy for therapy refractory extra-adrenal Cushing's syndrome.

PURPOSE: In selected cases of severe Cushing's syndrome due to uncontrolled ACTH secretion, bilateral adrenalectomy appears unavoidable. Compared with unilateral adrenalectomy (for adrenal Cushing's syndrome), bilateral adrenalectomy has a perceived higher perioperative morbidity. The aim of the current study was to compare both interventions in endogenous Cushing's syndrome regarding postoperative outcomes. METHODS: We report a single-center, retrospective cohort study comparing patients with hypercortisolism undergoing bilateral vs. unilateral adrenalectomy during 2008-2021. Patients with adrenal Cushing's syndrome due to adenoma were compared with patients with ACTH-dependent Cushing's syndrome (Cushing's disease and ectopic ACTH production) focusing on postoperative morbidity and mortality as well as long-term survival. RESULTS: Of 83 patients with adrenalectomy for hypercortisolism (65.1% female, median age 53 years), the indication for adrenalectomy was due to adrenal Cushing's syndrome in 60 patients (72.2%; 59 unilateral and one bilateral), and due to hypercortisolism caused by Cushing's disease (n = 16) or non-pituitary uncontrolled ACTH secretion of unknown origin (n = 7) (27.7% of all adrenalectomies). Compared with unilateral adrenalectomy (n = 59), patients with bilateral adrenalectomy (n = 24) had a higher rate of severe complications (0% vs. 33%; p < 0.001) and delayed recovery (median: 10.2% vs. 79.2%; p < 0.001). Using the MTL30 marker, patients with bilateral adrenalectomy fared worse than patients after unilateral surgery (MTL30 positive: 7.2% vs. 25.0% p < 0.001). Postoperative mortality was increased in patients with bilateral adrenalectomy (0% vs. 8.3%; p = 0.081). CONCLUSION: While unilateral adrenalectomy for adrenal Cushing's syndrome represents a safe and definitive therapeutic option, bilateral adrenalectomy to control ACTH-dependent extra-adrenal Cushing's syndrome or Cushing's disease is a more complicated intervention with a mortality of nearly 10%.

Types and Aspects of Front-of-Package Labeling Preferred by Parents: Insights for Policy Making in China.

The WHO recommends front-of-package labeling (FOPL) to help parents make healthier food choices for their children. But which type of FOPL resonates with parents in China? We performed a cross-sectional study to investigate parental preferences for five widely used formats of FOPL. A multi-stage cluster sampling method was applied to selected parents of students in primary and secondary schools in six provinces and municipalities from July 2020 to March 2021. A close-ended questionnaire was used to collect demographic information, parents' preferences for five FOPL in three dimensions, perceptions of the importance of nutrients labeled on FOPL, and prepackaged foods that need FOPL most. Chi-square tests were used to examine the characteristics among five groups. The results showed that multiple traffic lights (MTL) was preferred by parents, followed by warning labels. Parents thought the most needed nutrients to label were sugar, salt, and total fat. The top three prepackaged foods to label were "baked food", "milk and dairy products" and "sugar-sweetened beverages". Our findings indicate that nutrient-specific FOPL formats with interpretive aids were preferred by Chinese parents. These new findings can help inform the planning and implementation of FOPL in China and help Chinese parents make healthier food choices.

Influence of target-distractor neural similarity on working memory performance in older and younger adults.

According to the inhibitory deficit hypothesis, older adults often fail to selectively inhibit distractors and attend to relevant information in working memory, leading to poorer memory of target items but better recall of irrelevant distractors compared to younger adults. Here, we explored how neural similarity of activity patterns between relevant and irrelevant stimulus categories impacts memory performance. We found evidence that older adults may benefit from failing to inhibit distractors that are similar to targets, perhaps because sustained neural activation of distractors partially supports maintenance of targets when they share neural resources, allowing for better subsequent recognition of studied target items. We also found increased category-specific multivoxel pattern activity in medial temporal regions in younger compared to older adults as category similarity increased. We propose that this reduced category-specific activation in medial temporal regions in older adults may reflect more blended representations of all the information available in working memory.

Candida albicans MTLa2 regulates the mating response through both the a-factor and α-factor sensing pathways.

The diploid fungal pathogen Candida albicans has three configurations at the mating type locus (MTL): heterozygous (a/α) and homozygous (a/a or α/α). C. albicans MTL locus encodes four transcriptional regulators (MTLa1, a2, α1, and α2). The conserved a1/α2 heterodimer controls not only mating competency but also white-opaque heritable phenotypic switching. However, the regulatory roles of MTLa2 and α1 are more complex and remain to be investigated. MTLa/a cells often express a cell type-specific genes and mate as the a-type partner, whereas MTLα/α cells express α-specific genes and mate as the α-type partner. In this study, we report that the MTLa2 regulator controls the formation of mating projections through both the a- and α-pheromone-sensing pathways and thus results in the bi-mater feature of "α cells" of C. albicans. Ectopic expression of MTLa2 in opaque α cells activates the expression of not only MFA1 and STE3 (a-pheromone receptor) but also MFα1 and STE2 (α-pheromone receptor). Inactivation of either the MFa-Ste3 or MFα-Ste2 pheromone-sensing pathway cannot block the MTLa2-induced development of mating projections. However, the case is different in MTLα1-ectopically expressed opaque a cells. Inactivation of the MFα-Ste2 but not the MFa-Ste3 pheromone-sensing pathway blocks MTLα1-induced development of mating projections. Therefore, MTLa2 and MTLα1 exhibit distinct regulatory features that control the mating response in C. albicans. These findings shed new light on the regulatory mechanism of bi-mating behaviors and sexual reproduction in C. albicans.

Volume changes of medial temporal lobe structures in patients with genetic generalized and temporal lobe epilepsy in relation to neuropsychological functions.

Purpose: In patients with epilepsy (PWE), cognitive and behavioural dysfunctions are associated with abnormalities in various brain areas. The aim of the study was to compare the volume of the hippocampus (VHIP), amygdala (VAMG) and parahippocampal gyrus (VPHG) with the results of neuropsychological assessment in patients with temporal lobe epilepsy (TLE) and genetic generalized epilepsy (GGE). Methods: 33 PWE were enrolled in the study (mean age 37.3), 10 with TLE and 23 GGE (12 with GGE with tonic-clonic seizure [GGE-GTCS], and 11 with juvenile myoclonic epilepsy). 19 healthy persons (mean age 32.2) were enrolled as the control group (CG). Measurements of VHIP, VAMG and VPHG were made with 3D completely balanced steady state (CBASS) and 3D T1-weighted sequence. All participants underwent a neuropsychological assessment using a multi-domain cognitive battery and emotional state questionnaires. Results: The left hippocampus was significantly smaller in patients with left TLE (LTLE) and with GGE-GTCS, compared to the CG (p = 0.0069). In LTLE a significant enlargement of the right amygdala in comparison to the CG and other types of epilepsy were found (p = 0.0015). Among patients with LTLE and GGE-GTCS, impairment of attention and executive functions was statistically more common than in the CG. VHIP right (r = 0.25 p < 0.01) and VHIP left (r = 0.26 p < 0.04) were positively correlated with phonetic verbal fluency. Conclusions: PWE showed changes in the volume of selected medial temporal lobe (MTL) structures. Selective impairment of attention and executive functions was found. Some neuropsychological findings correlate with volume changes in MTL structures. Antiseizure medications therapy could have an impact on the severity of neuropsychological dysfunctions.

Early memory deficits and extensive brain network disorganization in the AppNL-F/MAPT double knock-in mouse model of familial Alzheimer's disease.

A critical challenge in current research on Alzheimer's disease (AD) is to clarify the relationship between network dysfunction and the emergence of subtle memory deficits in itspreclinical stage. The AppNL-F/MAPT double knock-in (dKI) model with humanized ß-amyloid peptide (Aß) and tau was used to investigate both memory and network dysfunctions at an early stage. Young male dKI mice (2 to 6 months) were tested in three tasks taxing different aspects of recognition memory affected in preclinical AD. An early deficit first appeared in the object-place association task at the age of 4 months, when increased levels of ß-CTF and Aß were detected in both the hippocampus and the medial temporal cortex, and tau pathology was found only in the medial temporal cortex. Object-place task-dependent c-Fos activation was then analyzed in 22 subregions across the medial prefrontal cortex, claustrum, retrosplenial cortex, and medial temporal lobe. Increased c-Fos activation was detected in the entorhinal cortex and the claustrum of dKI mice. During recall, network efficiency was reduced across cingulate regions with a major disruption of information flow through the retrosplenial cortex. Our findings suggest that early perirhinal-entorhinal pathology is associated with abnormal activity which may spread to downstream regions such as the claustrum, the medial prefrontal cortex and ultimately the key retrosplenial hub which relays information from frontal to temporal lobes. The similarity between our findings and those reported in preclinical stages of AD suggests that the AppNL-F/MAPT dKI model has a high potential for providing key insights into preclinical AD.