FreqSense: Adaptive Sampling Rates for Sensor-Based Human Activity Recognition Under Tunable Computational Budgets.

Recent years have witnessed great success of deep convolutional networks in sensor-based human activity recognition (HAR), yet their practical deployment remains a challenge due to the varying computational budgets required to obtain a reliable prediction. This article focuses on adaptive inference from a novel perspective of signal frequency, which is motivated by an intuition that low-frequency features are enough for recognizing "easy" activity samples, while only "hard" activity samples need temporally detailed information. We propose an adaptive resolution network by combining a simple subsampling strategy with conditional early-exit. Specifically, it is comprised of multiple subnetworks with different resolutions, where "easy" activity samples are first classified by lightweight subnetwork using the lowest sampling rate, while the subsequent subnetworks in higher resolution would be sequentially applied once the former one fails to reach a confidence threshold. Such dynamical decision process could adaptively select a proper sampling rate for each activity sample conditioned on an input if the budget varies, which will be terminated until enough confidence is obtained, hence avoiding excessive computations. Comprehensive experiments on four diverse HAR benchmark datasets demonstrate the effectiveness of our method in terms of accuracy-cost tradeoff. We benchmark the average latency on a real hardware.

ProtoHAR: Prototype Guided Personalized Federated Learning for Human Activity Recognition.

Federated Learning (FL) has recently attracted great interest in sensor-based human activity recognition (HAR) tasks. However, in real-world environment, sensor data on devices is non-independently and identically distributed (Non-IID), e.g., activity data recorded by most devices is sparse, and sensor data distribution for each client may be inconsistent. As a result, the traditional FL methods in the heterogeneous environment may incur a drifted global model that causes slow convergence and a heavy communication burden. Although some FL methods are gradually being applied to HAR, they are designed for overly ideal scenarios and do not address such Non-IID problem in the real-world setting. It is still a question whether they can be applied to cross-device FL. To tackle this challenge, we propose ProtoHAR, a prototype-guided FL framework for HAR, which aims to decouple the representation and classifier in the heterogeneous FL setting efficiently. It leverages the global prototype to correct the activity feature representation to make the prototype knowledge flow among clients without leaking privacy while solving a better classifier to avoid excessive drift of the local model in personalized training. Extensive experiments are conducted on four publicly available datasets: USC-HAD, UNIMIB-SHAR, PAMAP2, and HARBOX, which are collected in both controlled environments and real-world scenarios. The results show that compared with the state-of-the-art FL algorithms, ProtoHAR achieves the best performance and faster convergence speed in HAR datasets.

Protective effects of Paeoniflorin against AOPP-induced oxidative injury in HUVECs by blocking the ROS-HIF-1α/VEGF pathway.

BACKGROUND: Paeoniflorin, a monoterpene glycoside, exerts protective vascular effects, showing good antioxidant properties. However, whether Paeoniflorin has protective effect against the oxidative damage induced by advanced oxidation protein products (AOPPs) in Human umbilical vein endothelial cells (HUVECs) is unknown, as is the underlying mechanism. PURPOSE: The present study was designed to investigate the effect of Paeoniflorin on oxidative damage of HUVECs and elucidate its underlying molecular mechanisms. METHODS: The fluorescence intensity of 2', 7'-dichlorofluorescein-diacetate (DCFH-DA) staining was detected for intracellular reactive oxygen species (ROS) production. The increases mitochondrial membrane potential (MMP) was measured via flow cytometry and confocal microscopy using MitoTracker® Deep Red/ MitoTracker® Green staining. The intracellular adenosine triphosphate (ATP) was measured by ATP Determination Kit according to the manufacturer's protocol. Nox2, Nox4, hypoxia inducible factor-1α (HIF-1α), vascular endothelial growth factor (VEGF) and nuclear factor-κB (NF-κB) p65 expressions were detected by western blot. RESULTS: Our results showed that Paeoniflorin increases MMP and ATP levels of HUVECs induced by AOPPs, and attenuates NF-κB p65 expression on HUVECs might mainly result from its antioxidant capability by suppressing ROS production. Moreover, we also found that Paeoniflorin can suppress HIF-1α and VEGF protein expression through a decrease of ROS production via down-regulation of Nox2/Nox4 expression in HUVECs. AOPP-induced RAGE mRNA up-regulation was blocked by Paeoniflorin treatment in HUVECs. CONCLUSION: Our results provided the first experimental that Paeoniflorin protects against AOPP-induced oxidative damage in HUVECs, mainly through a mechanism involving a decrease in ROS production by the inhibition of Nox2/Nox4 and RAGE expression; restored ATP depletion and mitochondria dysfunction via ROS suppression; and down-regulated HIF-1α/VEGF, possibly via the ROS-NF-κB axis.

Huang Gan Formula Eliminates the Oxidative Stress Effects of Advanced Oxidation Protein Products on the Divergent Regulation of the Expression of AGEs Receptors via the JAK2/STAT3 Pathway.

Chronic kidney disease (CKD) has a high prevalence and low cure rate and represents a significant health issue. Oxidative stress is common in CKD due to metabolic disorders, inflammation, and impaired renal function changing normal proteins into advanced oxidation protein products (AOPPs). Huang Gan formula (HGF) is a new type of traditional Chinese herbal medicine. Although we previously investigated the protective effects of HGF against oxidative stress, the mechanism of HGF in CKD is still not fully understood. In this study, we used western blotting, quantitative polymerase chain reaction, and biochemical assays to show that HGF significantly decreased AOPP-induced oxidative stress damage. Moreover, the protective effects of HGF might be associated with upregulation of the advanced glycation end product receptor 1 (AGE-R1) and downregulation of the receptor for advance glycation end products (RAGE). Treatment with HGF and the Janus kinase 2 (JAK2) inhibitor, AG4-90, significantly attenuated AOPP-induced JAK2/STAT3 protein levels. These findings indicate that HGF inhibits AOPP-mediated biological responses by inactivating the JAK2/STAT3 pathway. In conclusion, HGF eliminated AOPP-induced effects in human mesangial cells (HMCs) by interrupting JAK2/STAT3 signaling, which altered RAGE/AGE-R1 expression and reduced oxidative stress in CKD.

Protective effect of Huang Gan formula in 5/6 nephrectomized rats by depressing the Wnt/ß-catenin signaling pathway.

Huang Gan formula (HGF) is a new traditional Chinese herbal medicine created according to the basic theory of traditional Chinese medicine. The aim of this study is to evaluate the effects of HGF on chronic kidney disease and determine the mechanisms of action. The extract of HGF was prepared, and qualitative and quantitative determination of phytochemical was performed with quadrupole time-of-flight mass spectrometer and high-performance liquid chromatography. Sprague-Dawley rats (n=72) were submitted to 5/6 nephrectomy (Nx), and then respectively treated with uremic clearance granule, losartan, HGF low dose, HGF middle dose, and HGF high dose once per day for 12 weeks. The sham group of operated rats (n=22) was treated with normal saline or HGF middle dose as a background control group. Blood and urine biochemical parameters, renal tissue morphology, and mRNA and proteins of Wnt/ß-catenin signaling pathways were investigated. The results showed that the quality of the extraction process could be controlled, and a total of eight major compounds were identified and quantified. HGF could decrease the level of serum creatinine, blood urea nitrogen, and urine protein and increase the renal index and creatinine clearance rate in a dose-dependent manner. HGF also remarkably reduced the glomerulosclerosis and tubulointerstitial fibrosis by blocking the Wnt/ß-catenin signaling pathway through inhibiting the Wnt1, ß-catenin, transcription factor 4, and fibronectin 1 expressions, simultaneously measured through mRNA and protein levels in the remnant kidney. These results suggest that extraction of HGF could improve remnant renal function and possibly ameliorate glomerulosclerosis and tubulointerstitial fibrosis by depressing the Wnt/ß-catenin signaling pathway.