Research on Convolutional Neural Network Inference Acceleration and Performance Optimization for Edge Intelligence.

In recent years, edge intelligence (EI) has emerged, combining edge computing with AI, and specifically deep learning, to run AI algorithms directly on edge devices. In practical applications, EI faces challenges related to computational power, power consumption, size, and cost, with the primary challenge being the trade-off between computational power and power consumption. This has rendered traditional computing platforms unsustainable, making heterogeneous parallel computing platforms a crucial pathway for implementing EI. In our research, we leveraged the Xilinx Zynq 7000 heterogeneous computing platform, employed high-level synthesis (HLS) for design, and implemented two different accelerators for LeNet-5 using loop unrolling and pipelining optimization techniques. The experimental results show that when running at a clock speed of 100 MHz, the PIPELINE accelerator, compared to the UNROLL accelerator, experiences an 8.09% increase in power consumption but speeds up by 14.972 times, making the PIPELINE accelerator superior in performance. Compared to the CPU, the PIPELINE accelerator reduces power consumption by 91.37% and speeds up by 70.387 times, while compared to the GPU, it reduces power consumption by 93.35%. This study provides two different optimization schemes for edge intelligence applications through design and experimentation and demonstrates the impact of different quantization methods on FPGA resource consumption. These experimental results can provide a reference for practical applications, thereby providing a reference hardware acceleration scheme for edge intelligence applications.

Association between air pollution and the prevalence of allergic rhinitis in Chinese children: A systematic review and meta-analysis.

Background: Allergic rhinitis (AR) is a common chronic inflammatory disease with bothersome symptoms. However, the effect of air pollution on the prevalence of AR in children is controversial. Objective: This study aimed to investigate the association between air pollution and the prevalence of AR in Chinese children. Methods: This study, in China, included 160,356 students ages 0-18 years who completed a questionnaire about the accuracy of the International Study of Asthma and Allergies in Childhood (ISAAC). The effect of different air pollutants on the prevalence rate were evaluated by meta-analysis. Also, it evaluated the effect of different air pollutants on the prevalence rate. Results: The differences in the effects of sulfur dioxide (SO2) exposure (combined odds ratio [ORcombined] 1.03 [95% confidence interval {CI}, 1.01-1.05]; p = 0.010) and nitrogen dioxide (NO2) exposure (ORcombined 1.11 [95% CI, 1.05-1.18]; p = 0.0006) on the risk of childhood AR was statistically significant. The effect of particulate matter with aerodynamic diameter of <10 µm (PM10) exposure on the risk of childhood AR was statistically significant (ORcombined 1.02 [95% CI, 1.01-1.03]; p < 0.001), the effect of particulate matter with aerodynamic diameter of <2.5 µm (PM2.5) exposure on the risk of childhood AR was statistically significant (ORcombined 1.15 [95% CI, 1.03-1.29]; p = 0.02), and the effect of ozone exposure on the risk of childhood AR was not statistically significant (ORcombined 0.98 [95% CI, 0.67-1.41]; p = 0.13). Conclusion: NO2, SO2, PM2.5, and PM10 were associated with the prevalence of AR in Chinese children. PM2.5 had the highest correlation with AR prevalence.

Highly Expressed Soluble Recombinant Anti-GFP VHHs in Escherichia coli via Optimized Signal Peptides, Strains, and Inducers.

Antigen-binding variable domains of the H chain of heavy-chain antibodies (VHHs), also known as nanobodies (Nbs), are of great interest in imaging technique, disease prevention, diagnosis, and therapy. High-level expression of soluble Nbs is very important for its industrial production. In this study, we optimized the expression system of anti-green fluorescent protein (GFP) VHHs with three different signal peptides (SPs), outer-membrane protein A (OmpA), pectate lyase B (PelB), and L-asparaginase II SP (L-AsPsII), in different Escherichia coli strains via isopropyl ß-D-thiogalactoside (IPTG) induction and auto-induction, respectively. The solubility of recombinant anti-GFP VHHs with PelB or OmpA was significantly enhanced to the same extent by IPTG induction and auto-induction in BL21 (DE3) E. coli strain and the maximum yield of target protein reached approximately 0.4 mg/l in a shake flask. The binding activity of recombinant anti-GFP VHHs was also confirmed to be retained by native-polyacrylamide gel electrophoresis (PAGE). These results suggest that SPs like OmpA and PelB could efficiently improve the recombinant anti-GFP VHH solubility without changing its bioactivity, providing a novel strategy to optimize the E. coli expression system of soluble VHHs, and lay the foundation for the industrial production of soluble recombinant anti-GFP VHHs and the research of other VHHs in the future.

Multiscale Local Enhancement Deep Convolutional Networks for the Automated 3D Segmentation of Gross Tumor Volumes in Nasopharyngeal Carcinoma: A Multi-Institutional Dataset Study.

Purpose: Accurate segmentation of gross target volume (GTV) from computed tomography (CT) images is a prerequisite in radiotherapy for nasopharyngeal carcinoma (NPC). However, this task is very challenging due to the low contrast at the boundary of the tumor and the great variety of sizes and morphologies of tumors between different stages. Meanwhile, the data source also seriously affect the results of segmentation. In this paper, we propose a novel three-dimensional (3D) automatic segmentation algorithm that adopts cascaded multiscale local enhancement of convolutional neural networks (CNNs) and conduct experiments on multi-institutional datasets to address the above problems. Materials and Methods: In this study, we retrospectively collected CT images of 257 NPC patients to test the performance of the proposed automatic segmentation model, and conducted experiments on two additional multi-institutional datasets. Our novel segmentation framework consists of three parts. First, the segmentation framework is based on a 3D Res-UNet backbone model that has excellent segmentation performance. Then, we adopt a multiscale dilated convolution block to enhance the receptive field and focus on the target area and boundary for segmentation improvement. Finally, a central localization cascade model for local enhancement is designed to concentrate on the GTV region for fine segmentation to improve the robustness. The Dice similarity coefficient (DSC), positive predictive value (PPV), sensitivity (SEN), average symmetric surface distance (ASSD) and 95% Hausdorff distance (HD95) are utilized as qualitative evaluation criteria to estimate the performance of our automated segmentation algorithm. Results: The experimental results show that compared with other state-of-the-art methods, our modified version 3D Res-UNet backbone has excellent performance and achieves the best results in terms of the quantitative metrics DSC, PPR, ASSD and HD95, which reached 74.49 ± 7.81%, 79.97 ± 13.90%, 1.49 ± 0.65 mm and 5.06 ± 3.30 mm, respectively. It should be noted that the receptive field enhancement mechanism and cascade architecture can have a great impact on the stable output of automatic segmentation results with high accuracy, which is critical for an algorithm. The final DSC, SEN, ASSD and HD95 values can be increased to 76.23 ± 6.45%, 79.14 ± 12.48%, 1.39 ± 5.44mm, 4.72 ± 3.04mm. In addition, the outcomes of multi-institution experiments demonstrate that our model is robust and generalizable and can achieve good performance through transfer learning. Conclusions: The proposed algorithm could accurately segment NPC in CT images from multi-institutional datasets and thereby may improve and facilitate clinical applications.

Geometric and Dosimetric Evaluation of Deep Learning-Based Automatic Delineation on CBCT-Synthesized CT and Planning CT for Breast Cancer Adaptive Radiotherapy: A Multi-Institutional Study.

PURPOSE: We developed a deep learning model to achieve automatic multitarget delineation on planning CT (pCT) and synthetic CT (sCT) images generated from cone-beam CT (CBCT) images. The geometric and dosimetric impact of the model was evaluated for breast cancer adaptive radiation therapy. METHODS: We retrospectively analyzed 1,127 patients treated with radiotherapy after breast-conserving surgery from two medical institutions. The CBCT images for patient setup acquired utilizing breath-hold guided by optical surface monitoring system were used to generate sCT with a generative adversarial network. Organs at risk (OARs), clinical target volume (CTV), and tumor bed (TB) were delineated automatically with a 3D U-Net model on pCT and sCT images. The geometric accuracy of the model was evaluated with metrics, including Dice similarity coefficient (DSC) and 95% Hausdorff distance (HD95). Dosimetric evaluation was performed by quick dose recalculation on sCT images relying on gamma analysis and dose-volume histogram (DVH) parameters. The relationship between ΔD95, ΔV95 and DSC-CTV was assessed to quantify the clinical impact of the geometric changes of CTV. RESULTS: The ranges of DSC and HD95 were 0.73-0.97 and 2.22-9.36 mm for pCT, 0.63-0.95 and 2.30-19.57 mm for sCT from institution A, 0.70-0.97 and 2.10-11.43 mm for pCT from institution B, respectively. The quality of sCT was excellent with an average mean absolute error (MAE) of 71.58 ± 8.78 HU. The mean gamma pass rate (3%/3 mm criterion) was 91.46 ± 4.63%. DSC-CTV down to 0.65 accounted for a variation of more than 6% of V95 and 3 Gy of D95. DSC-CTV up to 0.80 accounted for a variation of less than 4% of V95 and 2 Gy of D95. The mean ΔD90/ΔD95 of CTV and TB were less than 2Gy/4Gy, 4Gy/5Gy for all the patients. The cardiac dose difference in left breast cancer cases was larger than that in right breast cancer cases. CONCLUSIONS: The accurate multitarget delineation is achievable on pCT and sCT via deep learning. The results show that dose distribution needs to be considered to evaluate the clinical impact of geometric variations during breast cancer radiotherapy.

Multifunctional Bulk Hybrid Foam for Infrared Stealth, Thermal Insulation, and Microwave Absorption.

Taking serious microwave pollution issues and the complex application environment into consideration, it is quite urgent to integrate several functions into one material. Electromagnetic (EM) absorbing materials with multiple functions are highly attractive to next-generation wireless techniques and portable electronic devices. Herein, melamine foam provides a decent platform for the uniform growth of Co-based metal-organic frameworks (MOFs), which bring the as-obtained hybrid foam with three-dimensional porous network structure and combination of dielectric along with magnetic attenuation abilities as advanced materials in multifunctional fields. Remarkably, the relevant microwave absorption (MA) performance of the hybrid foam can reach an extremely high reflection loss value of -59.82 dB. Furthermore, the hybrid foam exhibits excellent infrared stealth and optimiztic heat insulation function, demonstrating the potential in plenty of practical applications. These results may arouse interests and inspirations of the elaborately design and facilely synthesis of high-performance foamlike microwave absorbers with multiple functions.

Affective Computing and the Impact of Gender and Age.

Affective computing aims at the detection of users' mental states, in particular, emotions and dispositions during human-computer interactions. Detection can be achieved by measuring multimodal signals, namely, speech, facial expressions and/or psychobiology. Over the past years, one major approach was to identify the best features for each signal using different classification methods. Although this is of high priority, other subject-specific variables should not be neglected. In our study, we analyzed the effect of gender, age, personality and gender roles on the extracted psychobiological features (derived from skin conductance level, facial electromyography and heart rate variability) as well as the influence on the classification results. In an experimental human-computer interaction, five different affective states with picture material from the International Affective Picture System and ULM pictures were induced. A total of 127 subjects participated in the study. Among all potentially influencing variables (gender has been reported to be influential), age was the only variable that correlated significantly with psychobiological responses. In summary, the conducted classification processes resulted in 20% classification accuracy differences according to age and gender, especially when comparing the neutral condition with four other affective states. We suggest taking age and gender specifically into account for future studies in affective computing, as these may lead to an improvement of emotion recognition accuracy.

Recognition of Intensive Valence and Arousal Affective States via Facial Electromyographic Activity in Young and Senior Adults.

BACKGROUND: Research suggests that interaction between humans and digital environments characterizes a form of companionship in addition to technical convenience. To this effect, humans have attempted to design computer systems able to demonstrably empathize with the human affective experience. Facial electromyography (EMG) is one such technique enabling machines to access to human affective states. Numerous studies have investigated the effects of valence emotions on facial EMG activity captured over the corrugator supercilii (frowning muscle) and zygomaticus major (smiling muscle). The arousal emotion, specifically, has not received much research attention, however. In the present study, we sought to identify intensive valence and arousal affective states via facial EMG activity. METHODS: Ten blocks of affective pictures were separated into five categories: neutral valence/low arousal (0VLA), positive valence/high arousal (PVHA), negative valence/high arousal (NVHA), positive valence/low arousal (PVLA), and negative valence/low arousal (NVLA), and the ability of each to elicit corresponding valence and arousal affective states was investigated at length. One hundred and thirteen participants were subjected to these stimuli and provided facial EMG. A set of 16 features based on the amplitude, frequency, predictability, and variability of signals was defined and classified using a support vector machine (SVM). RESULTS: We observed highly accurate classification rates based on the combined corrugator and zygomaticus EMG, ranging from 75.69% to 100.00% for the baseline and five affective states (0VLA, PVHA, PVLA, NVHA, and NVLA) in all individuals. There were significant differences in classification rate accuracy between senior and young adults, but there was no significant difference between female and male participants. CONCLUSION: Our research provides robust evidences for recognition of intensive valence and arousal affective states in young and senior adults. These findings contribute to the successful future application of facial EMG for identifying user affective states in human machine interaction (HMI) or companion robotic systems (CRS).

Similarities and differences of emotions in human-machine and human-human interactions: what kind of emotions are relevant for future companion systems?

Cognitive-technical intelligence is envisioned to be constantly available and capable of adapting to the user's emotions. However, the question is: what specific emotions should be reliably recognised by intelligent systems? Hence, in this study, we have attempted to identify similarities and differences of emotions between human-human (HHI) and human-machine interactions (HMI). We focused on what emotions in the experienced scenarios of HMI are retroactively reflected as compared with HHI. The sample consisted of N = 145 participants, who were divided into two groups. Positive and negative scenario descriptions of HMI and HHI were given by the first and second groups, respectively. Subsequently, the participants evaluated their respective scenarios with the help of 94 adjectives relating to emotions. The correlations between the occurrences of emotions in the HMI versus HHI were very high. The results do not support the statement that only a few emotions in HMI are relevant.