Fast Speed Research Articles

Association between subjective walking speed and metabolic diseases in individuals with obesity: a cross-sectional analysis.

The association between subjective walking speed and metabolic diseases has received limited attention, particularly in individuals with obesity. We aimed to clarify this association using comprehensive health checkup data of participants with obesity. In total, 8578 individuals with a body mass index ≥ 25.0kg/m2, 9626 individuals with waist circumference ≥ 85cm in men and ≥ 90cm in women, and 6742 individuals who met both criteria of body mass index and waist circumference were included in this cross-sectional analysis. Subjective walking speed was investigated using the question "Is your walking speed faster than the speed of those of your age and sex?" in a health examination questionnaire. Metabolic diseases were defined according to the guidelines for each disease, and modified Poisson regression analyses were performed. In the model adjusted for age and sex, individuals with obesity based on body mass index and fast subjective walking speed showed significantly lower risk of diabetes mellitus (risk ratio [RR] 0.70; 95% CI 0.63-0.77) and dyslipidemia (RR 0.97; 95% CI 0.94-1.00). Similarly, among those with obesity based on waist circumference and both body mass index and waist circumference, fast subjective walking speed showed a significant negative association with hypertension (RR 0.94; 95% CI 0.90-0.97 and RR 0.95; 95% CI 0.92-0.99, respectively), diabetes mellitus (RR 0.70; 95% CI 0.64-0.77 and RR 0.70; 95% CI 0.63-0.77, respectively), and dyslipidemia (RR 0.96; 95% CI 0.94-0.99 and RR 0.96; 95% CI 0.94-0.99, respectively). Thus, among individuals with obesity, the odds of metabolic diseases were lower if their subjective walking speed was fast. This study contributes to earlier prevention of the cascade of diseases that begin with obesity.

Comparative Analysis of Object Detection Models for Sheet Music Recognition: A Focus on YOLO and OMR Technologies

Abstract. As Artificial Intelligence (AI) technologies are developing rapidly and are widely used in various domains, it is efficient and convenient for composers to make music using AI to convert sheet music to audio. This research aims to compare the performance of different models in identifying individual notes within sheet music. Compared to traditional technologies like Optical Music Recognition (OMR), deep learning models have a significant advantage in processing blurry images with high efficiency. In the research process, three different models are used in searching for musical notes: OMR, You Only Look Once (YOLO)v5, and YOLOv8. The evaluation index consists of recognition accuracy, mean Average Precision (mAP), inference speed, and parameter quantity. After the experiment, it is found that the YOLO model performs best with high accuracy and fast speed. Based on the above analyses, the thesis finds that the YOLO model can be an efficient tool in composing music, with further research.

A Special Kind of Interplanetary Coronal Mass Ejections

Abstract It is generally believed that Coronal mass ejections (CMEs) have magnetic flux rope structures because of their helical shapes. However, only about 30-40% of interplanetary CMEs (ICMEs) have a local magnetic flux rope structure. The usually explanations are that the spacecraft only crossed the flank of the ropes and failed to detect the complete magnetic flux rope structure, or some processes destruct these magnetic flux rope structure. Several studies suggest that some ICMEs inherently possess disordered magnetic fields and consequently exhibit no magnetic flux-rope structures. We introduce a special kind of ICMEs. They have low magnetic field magnitude and stable magnetic field direction, relatively fast expansion speed, lower proton temperature and density. Their all three measured magnetic field components are relatively stable. We want to know whether these ICMEs also have magnetic flux rope structures or not. We identified 20 special ICMEs and analyzed their evolution based on their observed characteristics. We took a special ICME as example, which had apparent rope configuration at 1 AU but evolved to a special ICME at 5.4 AU, to illustrated that this kind of ICMEs may come from magnetic clouds (MCs) whose rope structure had been being stretched due to expansion. We inferred that the missing of obvious flux rope structure may be due to the expansion of MCs, not the flank crossing effect. However, more than 50% of the events were associated with dominant x-component of the magnetic filed, which indicates a leg crossing. Therefore, the detection of part of these special ICMEs may also be results of leg crossing effect.

Toward video-rate compressive spontaneous Raman imaging via single-photon avalanche diode arrays.

Spontaneous Raman microscopy is well-known for its remarkable chemical contrast yet suffers from slow acquisition speeds. Recently, the compressive Raman microspectroscopy framework has shown that a significant speed advantage is brought by leveraging shot-noise-limited detection using a single-photon avalanche diode (SPAD). However, current imaging speeds of compressive Raman architectures are fundamentally limited by SPAD sensitivity and dead time. Here, we demonstrate an efficient and scalable compressive Raman parallelization scheme based on SPAD arrays. We show that parallelization using line excitation, instead of spatial multiplexing, allows to reach effective pixel dwell times (τ pdt ) of 0.8 µs. Such fast speed represents over one order-of-magnitude speed-up over previous demonstrations. This effective parallelization not only allows for demonstrating unprecedented chemical imaging speeds using the otherwise weak spontaneous Raman effect but also paves the way for true video-rate inexpensive molecular microspectroscopy.

A review of research on optical true time delay technology

Light controlled phased array has the advantages of fast response speed, compact system, diverse functions, and flexible control, and has been widely applied in many scientific and technological fields. Optical true delay technology (OTTD) is the most direct technical means to achieve phase delay of optical carrier signals, and it is also the most basic technical means to implement optical controlled beamforming systems. In order to fully understand the optical true delay technology, this article first elaborates on the principle of phased array antennas and the reasons for beam strabismus, and analyzes the impact of true delay on the performance of phased array radar. Then, the basic principle, technological progress, and related applications of optical true delay are introduced. Taking four common structures of optical true delay lines as examples,which are micro-ring resonant cavity array, grating ture time delay line, multi-path switchable optical ture time delay line（OTTDL）, and wavelength selective optical delay line, their performance in delay accuracy, adjustable delay range, and frequency bandwidth are compared. Finally, the current problems and future development trends of optical controlled beamforming technology were summarized.

Optimized evaluation of the quality of sensor video internet of things (VIOT) by the integration of big data and artificial intelligence

AbstractThe application of sensor video internet of things technology to large-scale integrated work can significantly improve the working quality of employees. However, the degree of improvement in working quality is still difficult to measure in a systematic, intelligent, stable, and accurate manner. local optimization and adjustment after evaluation are still relatively challenging, To address these issues, the study proposes a method of optimizing the evaluation of sensor video quality through the integration of big data and AI techniques. A large-scale integrated distance education system in the field of education and training with a certain application basis is adopted as a case. Including big data and AI techniques such as integrated intelligent agent modules, recommendation algorithms, and transaction optimization algorithms, a new agent-oriented system design with fast response speed, strong scalability, convenient local optimization, and greater stability is achieved. According to the network topology structure of the distance education system in colleges and universities, this paper uses queuing theory to analyze the system performance of the system. The focus of this paper is the quantitative relationship between system communication intensity ρ, user arrival rate λ, system channel capacity n and system waiting delay, blocking probability, average queue length, system throughput and other important performance indicators. In teaching evaluation, the key factor that affects the quality of classroom teaching, that is, Developing a comprehensive system for evaluating classroom instruction is crucial. By incorporating student feedback, leveraging data mining techniques, and harnessing computer technology, a holistic framework for gathering, analyzing, and generating actionable insights on teaching performance is established. This approach makes the evaluation process more systematic and evidence-based, identifying 12 key elements that influence classroom education standards. In the experimental section, the student assessment data sets I1 and I2 exhibit experimental values (statistics) that significantly exceed the thresholds, with a minimum support of 0.32 and a confidence level of 0.61. Moreover, the Boolean matrix is divided into 90 points. The rule U1Ua ≥ U2 is identified as a subset of {U1U2Ua} within the large item set, signifying a strong association rule. These findings confirm the robustness of the artificial intelligence model proposed in this paper for video quality prediction. The optimized sensor video quality evaluation method not only meets a satisfactory confidence level and matching value but also demonstrates good reliability and relevance in the evaluation criteria.

Abstract 4134456: Hybrid Cardiac Targeting Peptide Ameliorates HFpEF Phenotype in Mice

Introduction: Heart failure with preserved ejection fraction (HFpEF) is a form of heart failure which is rapidly rising in incidence in the US today. Despite this, no therapies with mortality benefit exist for this syndrome. Previous work with our novel hybrid cardiac targeting peptide (hCTP), a cardiomyocyte-specific cell penetrating peptide, has shown multiple beneficial biological effects (improved calcium handling, anti-inflammatory properties) beyond its transduction abilities. Therefore, we hypothesized that treatment with hCTP in a mouse model of HFpEF would ameliorate the pathophysiology of this disease. Methods: To establish a HFpEF model, 6-week-old, C57BL/6NJ mice were placed on a high-fat diet (HFD), 60% kCal from lard, and water with 0.5 g/L of an iNOS inhibitor (N-Nitro-L-arginine methylester) to induce hypertension. After 5 weeks of the diet, HFD mice and their littermate controls underwent baseline exercise testing and echocardiograms. Exercise capacity was measured after 2 days of exercise training with 2 days of rest in between; mice were placed on a treadmill at a 10-degree angle and made to run at increasingly fast speeds (5-30 m/min) until exhaustion. Echocardiograms were performed under 1.5% inhalational isoflurane. After baseline testing, the mice were treated with vehicle, or either 2.5 or 5 mg/Kg hCTP, either 3 or 5 days a week (five groups, n = 3 per group). After four weeks of treatment, the mice were again subjected to the same battery of tests and euthanized. Data was compared using two way ANOVA with Sidak’s test to compare measurements pre- and post-treatment. Results: Treatment with hCTP at either concentration for five days/week led to an increase in running distance from 39.8 m to 116.3 m (p = 0.03, n = 6), compared to changes in control distance from 208.4 m to 157.8 m (p = 0.399). Running distance did not correlate with weight (R 2 < .001). Treatment with 5 mg/Kg hCTP for 5 days/week led to a normalization of the E/e’ ratio, from 43.3 to 26.9 (p = 0.004, n = 3) compared to a change in the control of 34.7 to 26.1 (p = 0.312). Ejection fractions remained consistent (>55%) across all groups. Conclusions: Our data suggests that hCTP appears to be a promising treatment in ameliorating some of the physiological effects in a murine model of HFpEF. Gene and protein expression studies in heart tissue are underway. Further work is warranted with a larger sample size.

A Fast Converter Synchronization Speed Method Based on the Frequency Difference on Both Sides of the Grid Connection Point

The challenge of achieving a reliable and safe synchronization process for microgrids under weak communication conditions is a significant issue in distributed grid-connected energy storage. This is also the core motivation of this study. First, the concept of weak communication is introduced, and weak communication conditions are simulated by limiting the number of communications. Additionally, a fast synchronization method based on the frequency difference at the grid connection point is proposed, which allows for the rapid synchronization of converters under weak communication conditions. This method also includes an analysis of the range for key parameters, providing practical and feasible guidance for its real-world application. Finally, the validity of the theory is verified through PSCAD/EMTDC simulations and physical experiments.

Are subtitling norms evidence-based? A narrative comparison of guidelines and subtitling advice with empirical evidence

ABSTRACT Despite comprehensive descriptions of norms in the literature, rapid evolution of subtitling practices often renders these outdated, and empirical investigation remains limited. This narrative scoping review aimed to provide a snapshot of current practice by investigating the consistency of English interlingual subtitling guidelines from four media providers (Netflix, Disney, TED Conferences, and Channel 4) and how these compare to academic prescriptions for practice. It also explored the extent to which current norms are supported by empirical evidence. Results compared instructions from media providers with academic recommendations and empirical research data in six key areas: (1) subtitle speed; (2) duration; (3) length; (4) segmentation; (5) position; and (6) appearance (color, font). Results indicated current guidelines are largely consistent across providers and allow faster speeds and longer lengths than previous norms. Additionally, results indicated subtitling norms have generally been proposed with little empirical justification, and empirical research conducted thus far has had limited generalizability. Going forward, a strong evidentiary basis for subtitling practices should be established to avoid unjustified restrictions on decision-making and to strengthen the professionalization of subtitling. Future research should explore impacts of various subtitling guidelines on audience comprehension and enjoyment to ensure that practices remain relevant and effective.

SPATIAL MARKET INTEGRATION AND PRICE TRANSMISSION OF COWPEA (Vigna unguiculata) IN NIGERIA

The study investigated spatial market integration and price transmission of cowpea in Nigeria. Secondary data on monthly basis for wholesale cowpea price per kilogram were used for the study. Augmented Dicky Fuller Test, Johansen Co-integration model, Granger Causality Test and Variance Decomposition of Vecto Autoregressive Model and Error Correction Model were used as analytical tools. Augmented Dickey Fuller test revealed that first difference all the price series were stationary and lag one (1) is the optimum lag length appropriate for the specified variables. Cointegration test results of the markets proved that despite that these markets were spatially separated geographically, they were connected in terms of price transmission across them, both in the long and short run but the level of integration is weak. It was observed that price changes are temporary and would converge to an equilibrium within a given time span. Granger causality test results indicate that one pair showed bi-directional causation between Bauchi and Enugu States. Thirty two (32) links showed unidirectional and three (3) pairs such as Niger→Abia, Niger→Lagos and Sokoto→Cross River States pairs showed no causal relationship between them. The study showed that Gombe and River States cowpea price had a faster speed of adjustment to equilibrium than other states. It also revealed that that 1% increase in Jigawa State will result to 20.5% increase in Enugu state and 1 % increase in River State will lead to 40.6% and 64.8% increase in Gombe and Nasarawa States in the short run respectively. The study concluded that, there is integration between the cowpea prices of source and destination states, but the degree of integration is weak. It is recommended that market participants should adopt standardize trading, settlement practice and invest in technology (electronic trading platform) to reduce operational inefficiencies in different markets to support integration.

Encoding of global visual motion in the avian pretectum shifts from a bias for temporal-to-nasal selectivity to omnidirectional excitation across speeds.

The pretectum of vertebrates contains neurons responsive to global visual motion. These signals are sent to the cerebellum, forming a subcortical pathway for processing optic flow. Global motion neurons exhibit selectivity for both direction and speed, but this is usually assessed by first determining direction preference at intermediate velocity (16-32 deg/sec), and then assessing speed tuning at the preferred direction. A consequence of this approach is that it is unknown if and how direction preference changes with speed. We measured directional selectivity in 114 pretectal neurons from 44 zebra finches (Taeniopygia guttata) across spatial and temporal frequencies, corresponding to a speed range of 0.062 to 1024°/s. Pretectal neurons were most responsive at 32-64°/s with lower activity as speed increased or decreased. At each speed, we determined if cells were directionally-selective, bidirectionally-selective, omnidirectionally responsive, or unmodulated. Notably, at 32°/s, 60% of the cells were directionally selective and 28% were omnidirectionally responsive. In contrast, at 1024°/s, 20% of the cells were directionally selective and nearly half of the population was omnidirectionally responsive. Only 15% of the cells were omnidirectionally excited across most speeds. The remaining 85% of the cells had direction tuning that changed with speed. Collectively, these results indicate a shift from a bias for directional tuning at intermediate speeds of global visual motion to a bias for omnidirectional responses at faster speeds. These results suggest a potential role for the pretectum during flight by detecting unexpected drift or potentials collisions, depending on the speed of the optic flow signal.Significance Statement During locomotion, images of edges and surfaces in the environment move across the retina, a signal of global visual motion called optic flow. Retinal recipient areas in the accessory optic system and the pretectum are the earliest sites to encode this signal, and the neurons are selective for direction and speed. Previous work suggested that directional selectivity may change across speeds but this has never been systematically studied. We measured direction preferences from 0.062 to 1024°/s in the avian pretectum. We found that pretectal global motion neurons are biased for temporal-to-nasal motion at intermediate speeds but biased for omnidirectional responses at faster speeds. These results suggest the pretectum could function to detect both unexpected drift and potential collisions during locomotion.

A new linearized ADI compact difference method on graded meshes for a nonlinear 2D and 3D PIDE with a WSK

In this work, a new linearized alternating direction implicit (ADI) compact difference method (CDM) is proposed for solving nonlinear two-dimensional (2D) and three-dimensional (3D) partial integrodifferential equation (PIDE) with a weakly singular kernel (WSK). The time derivative is treated by Crank-Nicolson (CN) method and the Riemann-Liouville (R-L) integral by product integration (PI) rule on graded meshes. The linear interpolation combining with Taylor formula is applied in time to deal with nonlinear term v∇v in interval (t0,t1), and linear interpolation concerning two previous time points is employed to deal with v∇v in intervals (tn−1,tn),n≥2. A linearized semi-discrete scheme is obtained, which can achieve second-order convergence in time. Then via introducing two kinds of compact difference operators to discretize the spatial derivatives. To improve the computing efficiency, we construct a ADI compact difference method. It is the first time that the ADI compact difference method is applied for the nonlinear 2D and 3D PIDE with a WSK. The advantage of our proposed scheme is that it not only has second-order accuracy in time and fourth-order accuracy in space, but also fast computational speed, just by solving the linear coupled equations for tridiagonal matrices. In addition, we prove the existence, uniqueness and convergence of 2D scheme. Four numerical examples in 2D and 3D are present to demonstrate our proposed method.

Screening, Identification and Application of Lactic Acid Bacteria for Degrading Mycotoxin Isolated from the Rumen of Yaks

Mycotoxin contamination is a major food safety issue worldwide, posing a serious threat to animal production performance and human health. Lactic acid bacteria are generally regarded as safe fermentation potential probiotics. They have the advantages of low toxicity, small pollution, strong specificity and high safety, and can reduce the contamination of microorganisms and mycotoxins. In this study, we compared the mycotoxin degradation capacity of 15 lactic acid bacteria strains from the rumen of Qinghai yak, through comprehensive analysis, we finally identified the strains as potential probiotics because they have a fast growth speed, strong acid production capacity (pH < 4.5), and they can grow normally in an environment with a pH of 3.5, bile salt concentration of 0.1%, and good self-agglutination and hydrophobicity. It was found that the fermentation group (Pediococcus acidilactici C2, Pediococcus acidilactici E28, Pediococcus pentosaceus A16 and Enterococcus lactis C16) could significantly reduce mycotoxin content, and both the nutritional and fermentation quality of the feed improved after 7 days of fermentation, meaning that the fermentation group can be used as a functional feed additive.

Research on rotational rollback suppression of diamond-shaped active locking flexure mechanism (ALFM) driven by piezoelectric wafers

Abstract Piezoelectric actuators have a wide range of applications in many areas due to their advantages of fast response speed, high resolution, compact and simple structure, diverse configurations, and resistance to electromagnetic interference. However, existing piezoelectric actuators generally have large fallback, and a few researchers have applied the ALFM to the fallback suppression of actuators. In this paper, an ALFM using a piezoelectric wafer as the driving source is innovatively designed, and the structure and dimensions are simulated and optimized using Finite Element Method, according to which the piezoelectric wafer-type ALFM is used to design a new inertial piezoelectric actuator that can suppress backward movement. The motion principle of the piezoelectric actuator is theoretically analyzed, followed by the establishment of the machine dynamics model of the piezoelectric actuator and simulation with MATLAB/simulink to verify the reasonableness of the dynamics model. Finally, a series of experiments are carried out on the processed drive model, and the results show that the maximum accuracy of the actuator is 8.5 μrad, and the maximum load capacity is 160 g. The comparison experiments at 30 Hz and 40 Hz with and without the ALFM prove that the locking mechanism does supress the actuator from backing off to a certain extent, which verifies the reasonableness of the scheme proposed in this paper.

LG-YOLOv8: A Lightweight Safety Helmet Detection Algorithm Combined with Feature Enhancement

In the realm of construction site monitoring, ensuring the proper use of safety helmets is crucial. Addressing the issues of high parameter values and sluggish detection speed in current safety helmet detection algorithms, a feature-enhanced lightweight algorithm, LG-YOLOv8, was introduced. Firstly, we introduce C2f-GhostDynamicConv as a powerful tool. This module enhances feature extraction to represent safety helmet wearing features, aiming to improve the efficiency of computing resource utilization. Secondly, the Bi-directional Feature Pyramid (BiFPN) was employed to further enrich the feature information, integrating feature maps from various levels to achieve more comprehensive semantic information. Finally, to enhance the training speed of the model and achieve a more lightweight outcome, we introduce a novel lightweight asymmetric detection head (LADH-Head) to optimize the original YOLOv8-n’s detection head. Evaluations on the SWHD dataset confirm the effectiveness of the LG-YOLOv8 algorithm. Compared to the original YOLOv8-n algorithm, our approach achieves a mean Average Precision (mAP) of 94.1%, a 59.8% reduction in parameters, a 54.3% decrease in FLOPs, a 44.2% increase in FPS, and a 2.7 MB compression of the model size. Therefore, LG-YOLOv8 has high accuracy and fast detection speed for safety helmet detection, which realizes real-time accurate detection of safety helmets and an ideal lightweight effect.

Improved adaptive robust control of direct-drive pump-valve cooperative brake-by-wire unit with disturbance compensation

Abstract Aiming at the needs of high-level autonomous driving, a direct-drive pump-valve coordinated brake-by-wire unit was designed. To improve the response speed, control accuracy, and robustness of the brake-by-wire unit, an improved adaptive integral robust control method (AIRC-AD) based on a novel adaptive law and disturbance compensation was proposed. By integrating pressure tracking error information and overall parameter estimation error information, a novel parameter adaptive law with fast convergence speed was designed. A finite-time disturbance observer was designed to observe the overall system disturbance, improving the system’s anti-disturbance performance. Finally, the stability of the control method was proven using the Lyapunov stability proof method. The results show that under step target signals, sinusoidal target signals and triangular wave target signals, AIRC-AD has better response speed and control accuracy. In conditions such as increased permanent magnet temperature, increased coil resistance, or reduced flow area due to hydraulic pipeline blockage, AIRC-AD has better adaptability. The unit and the method can meet the demand for precise braking force control in high-level autonomous driving.

A Horizontal Double‐Sided Copper Metallization Technology Designed for Solar Cell Mass‐Production

ABSTRACTIn the international photovoltaic market, the “carbon footprint” label has received great attention, and low‐carbon footprint products are widely popular. Compared with metallic silver, copper has the characteristics of low carbon emissions and low cost, which can effectively reduce the carbon footprint. Based on this, this article reports a horizontal double‐sided copper metallization technology. This technology can not only metalize the front and back sides of various types of silicon solar cells at the same time but also has fast speed, good uniformity, and simple process, making it suitable for the industrial mass production of solar cells. This article describes the structure and manufacturing process of TOPCon solar cells patterned with an ultrashort pulse laser and metalized using this novel horizontal double‐sided copper metallization technology. In this batch, average efficiency reached above 26%.

Scalable SOA-based Banyan optical space switch on InP membrane on silicon (IMOS)

Optical switches (OS) are vital for meeting modern data centers’ capacity and latency needs, overcoming the bandwidth and speed limitations that electronic switches encounter. To this end, this work proposes a semiconductor optical amplifier (SOA)-based OS on the IMOS, a unique platform with a combination of high refractive index contrast for compactness and monolithic active-passive integration for active functionalities, thereby enabling the creation of compact SOA-based OSes. An 8 × 8 Banyan SOA-based OS is integrated on a 4 × 4 mm2 area on the standard 4 × 6 mm2 IMOS cell. This marks the first application of the IMOS technology platform for SOA-based OSes and sets the stage for compact and large-scale monolithic photonic integrated OS circuits. The basic 2 × 2 switch module delivers a high optical signal-to-noise ratio (OSNR) and an extinction ratio (ER) above 45 dB. Employing NRZ-OOK routing on the 2 × 2 basic OS module, a 15 dB input power dynamic range (IPDR) is achieved within a 1 dB power penalty at 12.5 Gb/s. Data routing at higher data rates of 25 Gb/s and 40 Gb/s incurs power penalties of 0.8 and 1.4 dB, respectively. Furthermore, data routing for a 3-stage 8 × 8 switch operating at 25 Gb/s results in a power penalty of 1.2 dB. Compared to the 1-stage 2 × 2 switch, only a 0.4 dB additional penalty is observed despite incorporating 2 additional SOAs in the cascade. This indicates that the switch can scale to higher radix to accommodate multi-stage switches with numerous ports, large bandwidth, and fast speed, which is essential in modern large-scale data centers.

Comparative analysis of X-ray image classification of pneumonia based on deep learning algorithm

Abstract. In this paper, we compare and analyse the effectiveness of deep learning models such as AlexNet, Vgg, GoogleNet and MobileNet in the task of pneumonia image classification. During the training process, GoogleNet showed the fastest convergence speed, reaching convergence at the 3rd epoch; subsequently, the other three models gradually stabilised. In the end, the loss of AlexNet is 0.426, the loss of Vgg is 0.566, the loss of GoogleNet is 0.936, and the loss of MobileNet is 0.626. By selecting the weights of the round with the best training effect of each model, the results of classification accuracy are obtained: 88.9% for AlexNet, 92.6% for Vgg, and 92.6% for GoogleNet. 92.6%, GoogleNet is 85.2%, and MobileNet is 96.3%. MobileNet demonstrated the best prediction performance on the test set. These results provide a useful reference for the application of deep learning models in medical imaging..

Construction of Smart Tourism System Integrating Tourist Needs and Scene Characteristics

The demand for travel is increasing as human living conditions rise. The paper presents a smart tourism system architecture that incorporates visitors' demands and scenario characteristics, and performs path planning using path search algorithms and selective tour path recommendation algorithms, in order to improve tourists' travelling experiences and save them time. The experimental data showed that the enhanced heuristic search algorithm visited 122 nodes, which is 62.9% and 52.3% less than the sparrow search algorithm and the improved genetic search strategy, respectively. The number of iterations required to reach convergence for the selective tour path recommendation algorithm, genetic algorithm, discrete particle swarm algorithm, and genetic particle swarm algorithm, respectively, was 39, 90, 85, and 63, indicating that the proposed selective tour path recommendation algorithm has the fastest computational speed. The accuracy, stability, user satisfaction, and overall rating of the smart tourism system that integrates tourists' needs and scenario characteristics are all higher than those of the three types of tourism systems, such as the iBeacon Smart Tourism System, indicating that this smart tourism system is the best to use, helping to enhance tourists' experiences and promote the robust development of the tourism industry.