Step Size Research Articles

How do red foxes (Vulpes vulpes) explore their environment? Characteristics of movement patterns in time and space

Abstract Background Many animals must adapt their movements to different conditions encountered during different life phases, such as when exploring extraterritorial areas for dispersal, foraging or breeding. To better understand how animals move in different movement phases, we asked whether movement patterns differ between one way directed movements, such as during the transient phase of dispersal or two way exploratory-like movements such as during extraterritorial excursions or stationary movements. Methods We GPS collared red foxes in a rural area in southern Germany between 2020 and 2023. Using a random forest model, we analyzed different movement parameters, habitat features—for example landclasses and distances to linear structures—and time variables (season and time of day) within red fox exploratory, transient and stationary movement phases to characterize phase specific movement patterns and to investigate the influence of different variables on classifying the movement phases. Results According to the classification model, the movement patterns in the different phases were characterized most strongly by the variables persistence velocity, season, step length and distance to linear structures. In extraterritorial areas, red foxes either moved straight with high persistence velocity, close to anthropogenic linear structures during transient movements, or more tortuously containing a higher variance in turning angles and a decrease in persistence velocity during exploratory-like movements. Transient movements mainly took place during autumn, whereas exploratory-like movements were mainly conducted during winter and spring. Conclusion Movement patterns of red foxes differ between transient, exploratory and stationary phases, reflecting displacement, searching and resident movement strategies. Our results signify the importance of the combined effect of using movement, habitat and time variables together in analyzing movement phases. High movement variability may allow red foxes to navigate in extraterritorial areas efficiently and to adapt to different environmental and behavioral conditions.

Age-specific effects of a sustained cognitive activity on perceived cognitive fatigue as well as single- and dual-task treadmill walking performance.

During their daily lives humans are often confronted with sustained cognitive activities (SCA) leading to state fatigue, a psychobiological state characterized by a decrease in cognitive and/or motor performance and/or an increase in perception of fatigue. It was recently shown that performing SCA can impair overground dual-task gait performance in older adults, but it is currently unknown whether there is a task- and/or age-specific modulation in gait performance during treadmill walking. Therefore, the effect of a SCA on single- and dual-task treadmill walking performance was investigated in young and old adults. Using a crossover design, spatio-temporal gait parameters of 24 young and 23 older healthy participants were measured using motion capturing during single- and dual-task (including three cognitive interference tasks: word list generation, arithmetic, and Stroop-task) treadmill walking before and after SCA (30min Stroop-task) and a control task (reading). Moreover, cognitive fatigue, wakefulness, mood, and arousal were assessed. Although the SCA induced age-specific perceptual responses, no difference was found for cognitive performance during the Stroop-task. The cognitive interference task performance (word list generation, arithmetic, and Stroop-task) during walking on the treadmill did not decrease after the SCA. Single- and dual-task gait performance (e. g., step width and step length) specifically changed after the SCA and after the reading control task in both groups. Data indicate that perceived cognitive fatigue has an impact on single- and dual-task treadmill walking performance, with task- and age-specific differences. Although no general age-specific changes in single- and dual task gait performance following SCA were identified, perceived cognitive fatigue should be considered as an intrinsic risk factor for falls.

Thermal Image Classification of Autistic Children Using Res-Net Architecture

The thermal Image Classification Method has been widely used for significant applications in many fields, including thermal images of the face. This study presents a method for thermal facial classification in children with autism spectrum disorder (ASD). Children with ASD have a neurological disorder that affects communication skills essential in daily life and often causes difficulties in social situations. As we know, the diagnosis of ASD currently still relies on human methods and does not yet have definite biological markers. Early diagnosis of ASD has a significant positive impact, especially in children. Deep learning techniques, especially in facial medical image analysis, have become a new research focus in ASD detection. Initial screening using a Convolutional Neural Network (CNN) model with a transfer learning approach offers great potential for early diagnosis of ASD. The use of thermal imaging as a passive method to analyze ASD-related physiological signals has been proposed. In previous research, a deep learning model was developed to classify the faces of autistic children using thermal images. Therefore, this study aims to create a new Thermal Image Classification model for Autistic Children Using Res-Net Architecture. The architectures applied are ResNet-18, ResNet-34, and ResNet-50. As a comparison system, several of the same parameter values are used: epoch 100, batch size 2, SGD, Cross-entropy, learning rate 0.001, and momentum 0.9. The study test results show that the results of ResNet-18 are 97.22%, ResNet-34 99.22%, and ResNet-50 99.41%. Based on these results, ResNet-50 has the highest value.

Bilateral hip exoskeleton assistance enables faster walking in individuals with chronic stroke-related gait impairments

Millions of individuals surviving a stroke have lifelong gait impairments that reduce their personal independence and quality of life. Reduced walking speed is one of the major problems limiting community mobility and reintegration. Previous studies have shown positive effect of robot-assisted gait training utilizing hip exoskeletons for individuals with gait impairments due to a stroke, leading to increased walking speed in post-treatment compared to pre-treatment assessments. However, no evidence emerged of a significant increasing in walking speed attributable to device usage compared to walking without the device. In this pilot investigation, we observed that hip flexion/extension assistance delivered by a portable bilateral powered hip exoskeleton increased overground self-selected walking speed by 20.2 ± 5.0% on average among six chronic post-stroke survivors. When comparing walking with and without the hip exoskeleton within the same experimental session, the observed speed increment resulted in statistically and clinically meaningful improvement (0.14 ± 0.03 m/s > minimal clinically important difference, p = 0.015). The increased walking speed was the result of a higher self-selected cadence and longer step length both on the paretic and nonparetic limbs. By facilitating gait, a bilateral hip exoskeleton could be a viable technology for extending locomotor mobility and facilitating gait training of individuals affected by post-stroke hemiparesis.

User Behavior Analysis and Security Strategy Optimization of College English Learning Platform Using Dynamic Graph Convolutional Network

This paper aimed to address the shortcomings of current college English learning platforms in user behavior analysis and security protection, especially the lack of accurate abnormal behavior detection and dynamic identity authentication measures. By adopting DGCN (Dynamic Graph Convolutional Network) and XGBoost (eXtreme Gradient Boosting) classification models for abnormal behavior detection, a dynamic identity authentication optimization strategy based on behavior risk scoring is proposed to improve the security of the platform. The user embedding features output from DGCN are input into XGBoost for user behavior classification. Based on the predicted probability of the XGBoost classification model, a risk score is generated for each user, and different identity authentication strategies are defined according to the risk score range. Experimental results show that by modeling multi-layer dynamic graph structures, the accuracy, precision, recall and F1 score of the XGBoost model reached 97.6%, 97.5%, 95.3%, and 96.4% respectively. In addition, when the step size of the DGCN model increases from 1 to 10, its accuracy only slightly decreases from 97.6% to 97.5%, and the user behavior classification performance is stable. The incidence of account theft with dynamic identity authentication is maintained between 3% and 8%, and the average authentication time is increased by 2.1 seconds compared with the traditional password authentication method. The proposed DGCN can optimize the security strategy of the college English learning platform based on the user behavior risk score, and it can select targeted identity authentication strategies according to the user's risk score.

The acute effect of bitemporal electroconvulsive therapy on synchronous changes in heart rate variability and heart rate in patients with depression

Abstract Objective:
The transient autonomic nervous system responses induced by electroconvulsive therapy (ECT) may serve as critical indicators of treatment efficacy and potential side effects; however, their precise characteristics remains unclear. Considering that the intense stimulation of ECT may disrupt the typical antagonistic relationship between the sympathetic and parasympathetic branches, this study aims to conduct a meticulous analysis of the rapid changes in heart rate variability and heart rate during ECT, with a particular focus on their synchronized interplay.
Methods:
Pulse interval sequences were collected from fifty sessions of bitemporal ECT administered to twenty-seven patients diagnosed with major depressive disorder. The average heart rate (HR) and ultra-short term (UST) heart rate variability (HRV) indices RMSSD and SDNN, as well as the Poincaré indices SD1, SD2 and SD2/SD1, were calculated using a 10-second sliding window with a step size of 1 second. In particular, the synchronous changes between SD1, SD2, SD2/SD1 and HR were analyzed.
Results:
The synchronous changes of the indices showed different characteristics over time. In particular, SD1, SD2 and HR increased significantly by 41.50±11.45 ms, 33.97±10.98 ms and 9.68±2.00 bpm respectively between 8 and 20 seconds, whereas they decreased significantly by 19.89±9.07 ms, 17.54±8.54 ms and 3.80±1.33 bpm respectively between 45 and 53 seconds after ECT stimulus onset. SD1, SD2 and SD2/SD1 all had highly significant positive correlations with HR in the above phases.
Conclusion:
 The results suggest that bitemporal ECT induces the sympathetic and parasympathetic co-activation during the early ictal period and brief co-inhibition approximately 45 seconds after stimulus. Our findings may provide new insights comprehending the mechanisms of ECT and its associated cardiovascular risks.

ATKB-PID: an adaptive control method for micro tension under complex hot rolling conditions

At present, the parameters of the controllers in hot rolling roughing microtension control systems are not adaptively adjustable to variations in working conditions, which compromises both width accuracy and production stability. To address this issue, this paper introduces an ATKB-PID adaptive micro tension control method. This method incorporates a linear attention layer and utilizes a K-Nearest Neighbors (KNN) algorithm to predict the optimal learning rate and inertia coefficient under actual operating conditions. Furthermore, an objective function is tailored to production indices to enhance model performance. Comparative experiments with both established and recently introduced controllers demonstrate that the proposed ATKB-PID method exhibits a smaller steady-state error and quicker adjustment time. The ATKB-PID control method is well-suited for the complex and dynamic microtension control demands in thermal roughing processes, showing promising application potential.

Ankle muscle strength and activation are associated with walking patterns in preschool and school-age children.

Walking patterns can differ between children and adults, both kinematically and kinetically. However, the detailed nature of the ankle pattern has not been clarified. We investigated musculature, biomechanics, and muscle activation strategies and their relevance to walking performance in preschool (PS) and school children (SC), with adults (AD) as reference. Twenty-six PS (3-5 yr), 20 SC (6-8 yr), and 17 AD (18-30 yr) participated. Tibialis anterior (TA) and medial gastrocnemius (MG) thicknesses, fascicle lengths, and maximal voluntary dorsi- and plantar flexion isometric torques were measured. Hip, knee, and ankle kinematics, ground reaction forces (GRFs), and TA and MG electromyographic activities were recorded during shod walking at each participant's preferred speed. Walking speed, step length, and cadence were correlated with age in PS. These walking performance measures were also correlated with muscle thickness and fascicle length, showing higher speed in individuals with thicker muscles and longer TA and MG fascicles (conversely, higher cadence with thinner muscles and shorter fascicles). AD demonstrated the largest values for muscle thickness (p<0.001), fascicle length (p<0.001), strength (p<0.001), and walking performance (speed p = 0.004; step length p<0.001; cadence p<0.001), followed by SC then PS. Both PS and SC exhibited higher TA activities than AD during the stance phase, resulting in a higher co-activation index. The GRFs relative to body weight were lower in both horizontal and vertical components in PS compared to SC and AD, while the relative lateral force during stance was greatest in PS followed by SC and then AD. Differences in preferred walking speed and step length were associated with age and muscle size. Children, particularly preschool-aged, employed a co-activation strategy of dorsi- and plantar flexors for stabilization, which resulted in sideways steps even at a preferred walking speed.

Learning Rate Tuner with Relative Adaptation (LRT-RA): Road to Sustainable Computing

Optimizing learning rates (LRs) in deep learning (DL) has long been challenging. Previous solutions, such as learning rate scheduling (LRS) and adaptive learning rate (ALR) algorithms like RMSProp and Adam, added complexity by introducing new hyperparameters, thereby increasing the cost of model training through expensive cross-validation experiments. These methods mainly focus on local gradient patterns, which may not be effective in scenarios with multiple local optima near the global optimum. A new technique called Learning Rate Tuner with Relative Adaptation (LRT-RA) is introduced to tackle these issues. This approach dynamically adjusts LRs during training by analyzing the global loss curve, eliminating the need for costly initial LR estimation through cross-validation. This method reduces training expenses and carbon footprint and enhances training efficiency. It demonstrates promising results in preventing premature convergence, exhibiting inherent optimization behavior, and elucidating the correlation between dataset distribution and optimal LR selection. The proposed method achieves 84.96% accuracy on the CIFAR-10 dataset while reducing the power usage to 0.07 kWh, CO2 emissions to 0.05, and both SO2 and NOx emissions to 0.00003 pounds, during the whole training and testing process.

Pseudo-energy-preserving explicit Runge-Kutta methods

Using a recent characterization of energy-preserving B-series, we derive the explicit conditions on the coefficients of a Runge-Kutta method that ensure energy preservation (for Hamiltonian systems) up to a given order in the step size, which we refer to as the pseudo-energy-preserving (PEP) order. We study explicit Runge-Kutta methods with PEP order higher than their classical order. We provide examples of such methods up to PEP order six, and test them on Hamiltonian ODE and PDE systems. We find that these methods behave similarly to exactly energy-conservative methods over moderate time intervals and exhibit significantly smaller errors, relative to other Runge-Kutta methods of the same order, for moderately long-time simulations.

An insight from machine learning perspective for COVID-19 survival prediction in Malaysia based on demographic factor

Malaysia reported its first imported COVID-19 case on 23 January 2020, which marked the country’s first confirmed positive case. The first case in Malaysia was from eight close contacts in Johor. The global health landscape has been significantly impacted by the COVID-19 pandemic, with mortality or survival being critical outcomes of interest. This study aims to predict COVID-19 survival occurrences in Malaysia by utilizing machine learning approaches based on demographic factors. The dataset used in this study comprises demographic information of 2,151,315 COVID-19 patients, including nationality, regions, age groups, gender, medical history, vaccine brands, and the number of vaccine doses received between 2020 and 2022. Four machine learning algorithms, namely Logistic Regression, Naïve Bayes, Support Vector Machine, and Artificial Neural Network were employed to assess the relationship between demographic factors and COVID-19 survival. To evaluate the model performance, the datasets are categorized into imbalanced and balanced (down-sampling). The results indicate that the balanced dataset (down-sampling) outperforms the imbalanced dataset in terms of overall accuracy, sensitivity, specificity, precision, and Area Under the Curve (AUC). Based on the analysis, the Artificial Neural Network (ANN) classifier exhibited the highest performance with a specificity 95.2% on a balanced dataset. The model excels in accurately identifying survivors, thereby minimizing false mortality predictions and is selected as the best model for predicting COVID-19 survival. Its capacity to process larger sample sizes, combined with numerous interconnected nodes, enables it to identify complex patterns and extract meaningful insights from diverse datasets, such as demographic factors. Additionally, the optimization of parameters, including the number of layers, learning rate, and activation functions, significantly contributed to its superior accuracy. The study identifies that those of chronic diseases, male, and aged 45 and above as the notable factors associated with lower survival rates among COVID-19 patients. The findings underscore the importance of completing the vaccination series by obtaining at least the second dose, as the first dose alone may not offer sufficient protection. In conclusion, this study successfully achieves its objectives by identifying the optimal dataset configuration and predictive model for forecasting COVID-19 survival based on demographic factors. This network could serve as a benchmark model classifier, offering a valuable tool to predict and promote vaccinations, as well as optimize the general healthcare system during the pandemic outbreak. The study not only contributes to the theoretical understanding of effective COVID-19 prediction but also emphasizes the practical implications of integrating advanced machine learning techniques into pandemic management strategies. Future research can build upon these findings by exploring additional machine learning techniques and considering geographical and environmental factors to further enhance the accuracy of long-term predictions.

Deformation Detection Method for Substation Noise Barrier Column Based on Deep Learning and Digital Image Technology

The dynamic identification of the deformation of a noise barrier column is of great significance to the monitoring of its health. At the same time, the maximum stress of the column is an important indicator for the evaluation of its health status. Traditional contact displacement monitoring installs sensors on the structure, requires a lot of wiring and data acquisition equipment, and establishes a relatively independent and stable displacement reference system. Affected by the environment, wear, and material aging, the efficiency and reliability of data acquisition are reduced. A monitoring method based on digital image has the advantages of non-contact monitoring, high precision, and strong reliability. The existing DIC detection methods are limited by processor performance and image resolution, which are difficult to apply to engineering detection. In this paper, a structural displacement identification method based on convolutional neural networks (CNNs) and DIC technology is proposed. In this method, the data set is formed according to the column displacement cloud image obtained by DIC analysis, and the data set is enhanced by data normalization and region division. Through the analysis of the number of network layers and learning rate, the model design of the deep learning network is carried out. The high-speed camera image results of the test are introduced and identified by the static loading test of the equal-scale sound barrier. The results show that the structural displacement identification method based on CNN and DIC technology can accurately identify the displacement change in the structure, which greatly improves the efficiency of image displacement calculation using DIC technology.

Two‐Grid Finite Element Method for the Time‐Fractional Allen–Cahn Equation With the Logarithmic Potential

ABSTRACTIn this paper, we propose a two‐grid finite element method for solving the time‐fractional Allen–Cahn equation with the logarithmic potential. Firstly, with the L1 method to approximate Caputo fractional derivative, we solve the fully discrete time‐fractional Allen–Cahn equation on a coarse grid with mesh size and time step size . Then, we solve the linearized system with the nonlinear term replaced by the value of the first step on a fine grid with mesh size and the same time step size . We obtain the energy stability of the two‐grid finite element method and the optimal order of convergence of the two‐grid finite element method in the L2 norm when the mesh size satisfies . The theoretical results are confirmed by arithmetic examples, which indicate that the two‐grid finite element method can keep the same convergence rate and save the CPU time.

An Efficient 3D Convolutional Neural Network for Dose Prediction in Cancer Radiotherapy from CT Images

Introduction: Cancer is a highly lethal disease with a significantly high mortality rate. One of the most commonly used methods for treatment is radiation therapy. However, cancer treatment using radiotherapy is a time-consuming process that requires significant manual work from planners and doctors. In radiation therapy treatment planning, determining the dose distribution for each of the regions of the patient’s body is one of the most difficult and important tasks. Nowadays, artificial intelligence has shown promising results in improving the quality of disease treatment, particularly in cancer radiation therapy. Objectives: The main objective of this study is to build a high-performance deep learning model for predicting radiation therapy doses for cancer and to develop software to easily manipulate and use this model. Materials and Methods: In this paper, we propose a custom 3D convolutional neural network model with a U-Net-based architecture to automatically predict radiation doses during cancer radiation therapy from CT images. To ensure that the predicted doses do not have negative values, which are not valid for radiation doses, a rectified linear unit (ReLU) function is applied to the output to convert negative values to zero. Additionally, a proposed loss function based on a dose–volume histogram is used to train the model, ensuring that the predicted dose concentrations are highly meaningful in terms of radiation therapy. The model is developed using the OpenKBP challenge dataset, which consists of 200, 100, and 40 head and neck cancer patients for training, testing, and validation, respectively. Before the training phase, preprocessing and augmentation techniques, such as standardization, translation, and flipping, are applied to the training set. During the training phase, a cosine annealing scheduler is applied to update the learning rate. Results and Conclusions: Our model achieved strong performance, with a good DVH score (1.444 Gy) on the test dataset, compared to previous studies and state-of-the-art models. In addition, we developed software to display the dose maps predicted by the proposed model for each 2D slice in order to facilitate usage and observation. These results may help doctors in treating cancer with radiation therapy in terms of both time and effectiveness.

Additional chapter for evaluation indeterminate limits of functions and series in teaching mathematics for engineering education

Infinity divided by infinity, zero divided by zero, one divided by zero are the most important indeterminate forms obtained when evaluating limits for single variable functions and series. Well-known method; L' Hôpital rule has been employed to simplify and resolve these indeterminate forms' limits such that 0/0, ∞/∞ , 1/0 in terms of quotients of their derivatives. In some cases, L' Hôpital rule is applied more than once to solve indeterminate limits. Besides, it is so complicated to take the derivative for some functions of a single variable and series, so L' Hôpital rule is ineffective and not practical to solve limits with indeterminate forms for those functions and series. L' Hôpital rule is also impractical for the indeterminate limits in the form: - ∞∙0By considering all these facts, new approaches including Central Finite Difference (CFD), Forward Finite Difference (FFD), Backward Finite Difference (BFD), High Accurate Central Finite Difference (HACFD), High Accurate Forward Finite Difference (HAFFD), High Accurate Backward Finite Difference (HABFD) methods are presented that provides efficient ways to solve these limits. Taking derivative is not required in all approaches. HACFD with step size: 0.0001 is the most preferred technique here to obtain exact results among all other techniques.

COMPARING FORECASTS OF AGRICULTURAL SECTOR EXPORT VALUES USING SARIMA AND LONG SHORT-TERM MEMORY MODELS

Indonesia's agricultural sector plays a crucial role in the national economy, with significant export potential and supporting the livelihoods of the majority of the population. As part of the government's vision to make Indonesia the world's food barn by 2045, increasing the volume and value of agricultural product exports is a primary focus, making export value forecasting essential for supporting strategic decision-making. Sequential data analysis is an important approach in analyzing data collected over a specific period. This study aims to compare two popular methods in forecasting the export value of the agricultural sector, namely the Seasonal AutoRegressive Integrated Moving Average (SARIMA) model and the Long Short-Term Memory (LSTM) model. Monthly agricultural export data from West Java Province from January 2013 to February 2024 were used as the dataset. The best SARIMA model generated was (1,1,1)(0,1,1)12, while the optimal parameters for the LSTM model were neuron: 50, dropout rate: 0.3, number of layers: 2, activation function: relu, epochs: 500, batch size: 64, optimizer: Adam, and learning rate: 0.01. Evaluation was performed using the Root Mean Squared Error (RMSE) method to measure the accuracy of both models in forecasting the export value of the agricultural sector. The results showed that the LSTM model outperformed the SARIMA model, with a lower RMSE value (SARIMA: 4182.133 and LSTM: 1939.02). This research provides valuable insights for decision-makers in planning agricultural sector export strategies in the future. With this comparison, it is expected to provide better guidance in selecting forecasting methods suitable for the characteristics of the data.

Accelerating phase field simulations through a hybrid adaptive Fourier neural operator with U-net backbone

Prolonged contact between a corrosive liquid and metal alloys can cause progressive dealloying. For one such process as liquid-metal dealloying (LMD), phase field models have been developed to understand the mechanisms leading to complex morphologies. However, the LMD governing equations in these models often involve coupled non-linear partial differential equations (PDE), which are challenging to solve numerically. In particular, numerical stiffness in the PDEs requires an extremely refined time step size (on the order of 10−12s or smaller). This computational bottleneck is especially problematic when running LMD simulation until a late time horizon is required. This motivates the development of surrogate models capable of leaping forward in time, by skipping several consecutive time steps at-once. In this paper, we propose a U-shaped adaptive Fourier neural operator (U-AFNO), a machine learning (ML) based model inspired by recent advances in neural operator learning. U-AFNO employs U-Nets for extracting and reconstructing local features within the physical fields, and passes the latent space through a vision transformer (ViT) implemented in the Fourier space (AFNO). We use U-AFNOs to learn the dynamics of mapping the field at a current time step into a later time step. We also identify global quantities of interest (QoI) describing the corrosion process (e.g., the deformation of the liquid-metal interface, lost metal, etc.) and show that our proposed U-AFNO model is able to accurately predict the field dynamics, in spite of the chaotic nature of LMD. Most notably, our model reproduces the key microstructure statistics and QoIs with a level of accuracy on par with the high-fidelity numerical solver, while achieving a significant 11, 200 × speed-up on a high-resolution grid when comparing the computational expense per time step. Finally, we also investigate the opportunity of using hybrid simulations, in which we alternate forward leaps in time using the U-AFNO with high-fidelity time stepping. We demonstrate that while advantageous for some surrogate model design choices, our proposed U-AFNO model in fully auto-regressive settings consistently outperforms hybrid schemes.

Dynamic Control of Weight-Update Linearity in Magneto-Ionic Synapses.

Multifunctional hardware technologies for neuromorphic computing are essential for replicating the complexity of biological neural systems, thereby improving the performance of artificial synapses and neurons. Integrating ionic and spintronic technologies offers new degrees of freedom to modulate synaptic potentiation and depression, introducing novel magnetic functionalities alongside the established ionic analogue behavior. We demonstrate that magneto-ionic devices can perform as synaptic elements with dynamically tunable depression linearity controlled by an external magnetic field, a functionality reminiscent of neuromodulation in biological systems. By applying magnetic fields we significantly reduce the nonlinearity of synaptic depression, transitioning from an exponential dependence to a linear response at higher fields. Neural network simulations reveal that this magnetically induced linearity enhancement improves learning accuracy across a wide range of learning rates, which is retained after the magnetic field is removed. These findings highlight the versatility and promise of magneto-ionic devices for developing tunable synaptic elements for neuromorphic hardware.

ENHANCING 〖PM〗_(2.5) PREDICTION IN KEMAYORAN DISTRICT, DKI JAKARTA USING DEEP BILSTM METHOD

Worldwide air pollution is a concern, and this is especially true in Indonesia, where most people breathe air that is more contaminated than recommended by the WHO. The concentration of presents notable health hazards. The respiratory system is the primary route of absorption for , allowing it to enter the lung alveoli and enter the bloodstream. Given the significant health risks associated with exposure, accurate forecasting methods are crucial to anticipate and mitigate its effects. Traditional forecasting methods like ARIMA have limitations in handling non-linear and complex patterns. Therefore, an accurate machine learning method is needed to improve forecasting performance. This research employs Deep Bidirectional Long-Short Term Memory (BiLSTM), a deep learning model particularly suited for time series forecasting due to its ability to capture both past and future dependencies in sequential data. To achieve accurate and precise forecasts for predicting concentration levels in Kemayoran District in November , 2023 (24 hours), this research utilized hourly concentration data from May until October , 2023, using Deep BiLSTM. The outcomes demonstrated the efficiency of the model, attaining a Mean Absolute Percentage Error (MAPE) of 17.1540% (training) and 14.2862% (testing) with an 80:20 data split. The optimal parameters, which comprised 24 timesteps, Adam optimizers with a learning rate of 0.001, 16 batch sizes, 1000 epochs, and ReLU activation functions across multiple BiLSTM layers, showcased the model’s effectiveness in forecasting the concentration in Kemayoran District, DKI Jakarta, on November , 2023.

Gait Analysis in Patients After Hemiepiphysiodesis Due to Valgus or Varus Knee Deformity

Background: Developmental knee joint deformities are a common problem in pediatric orthopedics. Children with a valgus or varus deformity of the distal femur or the proximal tibia are commonly treated with hemiepiphysiodesis. Gait analysis in patients with lower limb deformities plays an important role in clinical practice. The purpose of our study was to assess gait parameters in patients who underwent hemiepiphysiodesis procedures of the distal femur or proximal tibia due to a knee deformity and to compare them with those in healthy controls. Methods: We prospectively evaluated 35 patients (14 females and 21 males) after hemiepiphysiodesis and compared the results with a healthy control group (26 participants). Gait was analyzed with a G-Sensor device (BTS Bioengineering Corp., Quincy, MA, USA). We assessed the following gait parameters: gait cycle duration, step length, support phase duration, swing phase duration, double support duration, single support duration, cadence, velocity, and step length. We assessed these gait parameters in a group of patients before and after treatment with hemiepiphysiodesis. We compared the patients’ results before and after treatment to those of a healthy control group. The level of significance was set at p &lt; 0.05. Results: The mean follow-up period was 13 months. There was no difference in the results of gait assessments in patients prior to and after treatment. The median step length was 47.09% in the treated limb after treatment and 54.01% in the intact limb (p = 0.018). There were no other differences in gait parameters in the treated limbs and the healthy, intact limbs in the patient group after treatment. There were no significant differences in the patients before and after treatment compared with those in the healthy control group in all gait parameters. Conclusions: Valgus or varus knee deformity correction with the use of hemiepiphysiodesis does not significantly improve preoperative gait parameters. The biomechanical outcomes of hemiepiphysiodesis in the treatment of valgus or varus knee deformity are good. We observed no differences in gait cycle duration, step length, support phase duration, swing phase duration, double support duration, single support duration, gait velocity, cadence, or step length between the experimental and healthy control groups.