Real-Time Sensor-Embedded Neural Network for Human Activity Recognition

This article introduces a novel approach to human activity recognition (HAR) by presenting a sensor that utilizes a real-time embedded neural network. The sensor incorporates a low-cost microcontroller and an inertial measurement unit (IMU), which is affixed to the subject's chest to capture their movements. Through the implementation of a convolutional neural network (CNN) on the microcontroller, the sensor is capable of detecting and predicting the wearer's activities in real-time, eliminating the need for external processing devices. The article provides a comprehensive description of the sensor and the methodology employed to achieve real-time prediction of subject behaviors. Experimental results demonstrate the accuracy and high inference performance of the proposed solution for real-time embedded activity recognition.

Neural correlates of working memory training in HIV patients: study protocol for a randomized controlled trial

Background

Potent combined antiretroviral therapy decreased the incidence and severity of HIV-associated neurocognitive disorders (HAND); however, no specific effective pharmacotherapy exists for HAND. Patients with HIV commonly have deficits in working memory and attention, which may negatively impact many other cognitive domains, leading to HAND. Since HAND may lead to loss of independence in activities of daily living and negative emotional well-being, and incur a high economic burden, effective treatments for HAND are urgently needed. This study aims to determine whether adaptive working memory training might improve cognitive functions and neural network efficiency and possibly decrease neuroinflammation. This study also aims to assess whether subjects with the LMX1A-rs4657412 TT(AA) genotype show greater training effects from working memory training than TC(AG) or CC(GG)-carriers.

Methods/Design

60 HIV-infected and 60 seronegative control participants will be randomized to a double-blind active-controlled study, using adaptive versus non-adaptive Cogmed Working Memory Training® (CWMT), 20–25 sessions over 5–8 weeks. Each subject will be assessed with near- and far-transfer cognitive tasks, self-reported mood and executive function questionnaires, and blood-oxygenation level-dependent functional MRI during working memory (n-back) and visual attention (ball tracking) tasks, at baseline, 1-month, and 6-months after CWMT. Furthermore, genotyping for LMX1A-rs4657412 will be performed to identify whether subjects with the TT(AA)-genotype show greater gain or neural efficiency after CWMT than those with other genotypes. Lastly, cerebrospinal fluid will be obtained before and after CWMT to explore changes in levels of inflammatory proteins (cytokines and chemokines) and monoamines.

Discussion

Improving working memory in HIV patients, using CWMT, might slow the progression or delay the onset of HAND. Observation of decreased brain activation or normalized neural networks, using fMRI, after CWMT would lead to a better understanding of how neural networks are modulated by CWMT. Moreover, validating the greater training gain in subjects with the LMX1A-TT(AA) genotype could lead to a personalized approach for future working memory training studies. Demonstrating and understanding the neural correlates of the efficacy of CWMT in HIV patients could lead to a safe adjunctive therapy for HAND, and possibly other brain disorders.

Trial registration

ClinicalTrial.gov, NCT02602418.

Schizophrenia-risk variant rs6994992 in the neuregulin-1 gene on brain developmental trajectories in typically developing children

The neuregulin-1 (NRG1) gene is one of the best-validated risk genes for schizophrenia, and psychotic and bipolar disorders. The rs6994992 variant in the NRG1 promoter (SNP8NRG243177) is associated with altered frontal and temporal brain macrostructures and/or altered white matter density and integrity in schizophrenic adults, as well as healthy adults and neonates. However, the ages when these changes begin and whether neuroimaging phenotypes are associated with cognitive performance are not fully understood. Therefore, we investigated the association of the rs6994992 variant on developmental trajectories of brain macro- and microstructures, and their relationship with cognitive performance. A total of 972 healthy children aged 3-20 years had the genotype available for the NRG1-rs6994992 variant, and were evaluated with magnetic resonance imaging (MRI) and neuropsychological tests. Age-by-NRG1-rs6994992 interactions and genotype effects were assessed using a general additive model regression methodology, covaried for scanner type, socioeconomic status, sex and genetic ancestry factors. Compared with the C-carriers, children with the TT-risk-alleles had subtle microscopic and macroscopic changes in brain development that emerge or reverse during adolescence, a period when many psychiatric disorders are manifested. TT-children at late adolescence showed a lower age-dependent forniceal volume and lower fractional anisotropy; however, both measures were associated with better episodic memory performance. To our knowledge, we provide the first multimodal imaging evidence that genetic variation in NRG1 is associated with age-related changes on brain development during typical childhood and adolescence, and delineated the altered patterns of development in multiple brain regions in children with the T-risk allele(s).

Fractone-heparan sulphates mediate FGF-2 stimulation of cell proliferation in the adult subventricular zone

OBJECTIVES

Fractones are extracellular matrix structures that form a niche for neural stem cells and their immediate progeny in the subventricular zone of the lateral ventricle (SVZa), the primary neurogenic zone in the adult brain. We have previously shown that heparan sulphates (HS) associated with fractones bind fibroblast growth factor-2 (FGF-2), a powerful mitotic growth factor in the SVZa. Here, our objective was to determine whether the binding of FGF-2 to fractone-HS is implicated in the mechanism leading to cell proliferation in the SVZa.

MATERIALS AND METHODS

Heparitinase-1 was intracerebroventricularly injected with FGF-2 to N-desulfate HS proteoglycans and determine whether the loss of HS and of FGF-2 binding to fractones modifies FGF-2 effect on cell proliferation. We also examined in vivo the binding of Alexa-Fluor-FGF-2 in relationship with the location of HS immunoreactivity in the SVZa.

RESULTS

Heparatinase-1 drastically reduced the stimulatory effect of FGF-2 on cell proliferation in the SVZa. Alexa-Fluor-FGF-2 binding was strictly co-localized with HS immunoreactivity in fractones and adjacent vascular basement membranes in the SVZa.

CONCLUSIONS

Our results demonstrate that FGF-2 requires HS to stimulate cell proliferation in the SVZa and suggest that HS associated with fractones and vascular basement membranes are responsible for activating FGF-2. Therefore, fractones and vascular basement membranes may function as a HS niche to drive cell proliferation in the adult neurogenic zone.