Seizure Detection and Prediction by Parallel Memristive Convolutional Neural Networks.

During the past two decades, epileptic seizure detection and prediction algorithms have evolved rapidly. However, despite significant performance improvements, their hardware implementation using conventional technologies, such as Complementary Metal-Oxide-Semiconductor (CMOS), in power and area-constrained settings remains a challenging task; especially when many recording channels are used. In this paper, we propose a novel low-latency parallel Convolutional Neural Network (CNN) architecture that has between 2-2,800x fewer network parameters compared to State-Of-The-Art (SOTA) CNN architectures and achieves 5-fold cross validation accuracy of 99.84% for epileptic seizure detection, and 99.01% and 97.54% for epileptic seizure prediction, when evaluated using the University of Bonn Electroencephalogram (EEG), CHB-MIT and SWEC-ETHZ seizure datasets, respectively. We subsequently implement our network onto analog crossbar arrays comprising Resistive Random-Access Memory (RRAM) devices, and provide a comprehensive benchmark by simulating, laying out, and determining hardware requirements of the CNN component of our system. We parallelize the execution of convolution layer kernels on separate analog crossbars to enable 2 orders of magnitude reduction in latency compared to SOTA hybrid Memristive-CMOS Deep Learning (DL) accelerators. Furthermore, we investigate the effects of non-idealities on our system and investigate Quantization Aware Training (QAT) to mitigate the performance degradation due to low Analog-to-Digital Converter (ADC)/Digital-to-Analog Converter (DAC) resolution. Finally, we propose a stuck weight offsetting methodology to mitigate performance degradation due to stuck [Formula: see text] memristor weights, recovering up to 32% accuracy, without requiring retraining. The CNN component of our platform is estimated to consume approximately 2.791 W of power while occupying an area of 31.255 mm2 in a 22 nm FDSOI CMOS process.

Aquaporin 1 alleviates acute kidney injury via PI3K-mediated macrophage M2 polarization.

BACKGROUND: Lipopolysaccharide (LPS)-induced acute kidney injury (AKI) is associated with an abnormal immune response. Accumulating evidence has demonstrated that aquaporin 1 (AQP1) prevents kidney tissue injury in LPS-induced AKI by mediating immune response. However, the underlying mechanisms remain obscure. Macrophages as immune cells with multiple phenotypes are important mediators in tissue homeostasis and host defense. We propose that macrophage polarization is implicated in AQP1-mediated immune response. METHODS: Herein we established sepsis-induced AKI model rats through intraperitoneal injection of LPS into Wistar rats to reveal immune mechanism of damage. We also used LPS-induced mouse RAW264.7 cells to elucidate the molecular mechanism of macropage polarization. RESULTS: Histopathology showed that renal tubular epithelial cells in the model group were swollen, inflammatory exudation was obvious and the inflammatory factors, interleukin-6 (IL-6) and tumor necrosis factor α (TNF-α) were increased. Western blotting showed PI3K was upregulated in the model group. Serum creatinine and urea nitrogen increased after LPS injection. Renal AQP1 mRNA is downregulated and serum AQP1 protein increased first and then decreased in LPS-induced AKI rats. M2 macrophage markers (Arg-1, CD206) were increased in repair stage. In addition, treatment of murine macrophages (RAW264.7) with AQP1 siRNA resulted in decreased PI3K activation and M2 polarization, but increased IL-6 and TNF-α. Moreover, inhibiting PI3K with wortmannin imitated the results of AQP1 silencing. CONCLUSIONS: Macrophage M2 polarization is likely the cellular mechanism underlying the anti-AKI property of AQP1, and PI3K activation is involved in the AQP1-induced M2 phenotype switch.

Aquaporin-1 attenuates macrophage-mediated inflammatory responses by inhibiting p38 mitogen-activated protein kinase activation in lipopolysaccharide-induced acute kidney injury.

OBJECTIVE: This study was designed to investigate the role of AQP1 in the development of LPS-induced AKI and its potential regulatory mechanisms in the inflammatory responses of macrophages. METHODS: Male Wistar rats were injected intraperitoneally with LPS, and biochemical and histological renal damage was assessed. The levels of inflammatory mediators, macrophage markers and AQP1 in blood and kidney tissues were assessed by ELISA. RTPCR was used to assess changes in the relative levels of AQP1 mRNA induced by LPS. Western blot and immunofluorescence analyses were performed to assay the activation of the p38 MAPK and NF-κB pathways, respectively. The same detection methods were used in vitro to determine the regulatory mechanisms underlying AQP1 function. RESULTS: AQP1 mRNA levels were dramatically decreased in AKI rats following the increased expression of inflammatory factors. In vitro experiments demonstrated that silencing the AQP1 gene increased inflammatory mediator secretion, altered the classical activation of macrophages, greatly enhanced the phosphorylation of p38 and accelerated the translocation of NF-κB. Furthermore, these results were blocked by doramapimod, a p38 inhibitor. Therefore, these effects were mediated by the increased phosphorylation of p38 MAPK. CONCLUSION: Our results suggest that altered AQP1 expression may be associated with the development of inflammation in AKI. AQP1 plays a protective role in modulating acute renal injury and can attenuate macrophage-mediated inflammatory responses by downregulating p38 MAPK activity in LPS-induced RAW264.7 cells. The pharmacological targeting of AQP1-mediated p38 MAPK signalling may provide a novel treatment approach for AKI.

Inhibition of inflammation using diacerein markedly improved renal function in endotoxemic acute kidney injured mice.

BACKGROUND: Inflammation is an important pathogenic component of endotoxemia-induced acute kidney injury (AKI), finally resulting in renal failure. Diacerein is an interleukin-1ß (IL-1ß) inhibitor used for osteoarthritis treatment by exerting anti-inflammatory effects. This study aims to investigate the effects of diacerein on endotoxemia-induced AKI. METHODS: Male C57BL/6 mice were intraperitoneally injected with lipopolysaccharide (LPS, 10 mg/kg) for 24 h prior to diacerein treatment (15 mg/kg/day) for another 48 h. Mice were examined by histological, molecular and biochemical approaches. RESULTS: LPS administration showed a time-dependent increase of IL-1ß expression and secretion in kidney tissues. Diacerein treatment normalized urine volume and osmolarity, reduced blood urea nitrogen (BUN), fractional excretion of sodium (FENa), serum creatinine and osmolarity, and protected renal function in an endotoxemic AKI mice model. In the histopathologic study, diacerein also improved renal tubular damage such as necrosis of the tubular segment. Moreover, diacerein inhibited LPS-induced increase of inflammatory cytokines, such as IL-1ß, tumor necrosis factor-α, monocyte chemoattractant protein-1 and nitric oxide synthase 2. In addition, LPS administration markedly decreased aquaporin 1 (AQP1), AQP2, AQP3, Na,K-ATPase α1, apical type 3 Na/H exchanger and Na-K-2Cl cotransporter expression in the kidney, which was reversed by diacerein treatment. We also found that diacerein or IL-1ß inhibition prevented the secretion of inflammatory cytokines and the decrease of AQP and sodium transporter expression induced by LPS in HK-2 cells. CONCLUSION: Our study demonstrates for the first time that diacerein improves renal function efficiently in endotoxemic AKI mice by suppressing inflammation and altering tubular water and sodium handing. These results suggest that diacerein may be a novel therapeutic agent for the treatment of endotoxemic AKI.