High Classification Research Articles

Analysis of Earthquake Vulnerability of The Demak Coastal Area based on the HVSR (Horizontal To Vertical Spectral Ratio) Method

<p>Most of the Demak Coastal area, especially on the north coast, has a subsurface structure which is quite thick sediment. The impact of an earthquake will cause damage to subsurface structures and buildings above it, so a method is needed that can determine earthquake vulnerability in an area to take mitigation steps. This research aims to analyze vulnerability to earthquakes based on natural frequency (fo), amplification (Ao) and seismic vulnerability index (SVI). The research was carried out by measuring the microseismic signal response at 89 locations using a 3-component seismograph and data logger. The research results showed that the dominant frequency varied from 0.26 – 5.26 Hz, the amplification factor varied from 0.51 – 3.56 and the seismic susceptibility index varied from 0.14-14.77 micro cm<sup>2</sup>/s. The study area was classified into low SVI (SVI < 5), medium SVI (5 < SVI < 10), and high SVI (SVI > 10). The potential earthquake hazard described by SVI with a value range of 10 < SVI (high classification) is found at observation station 60 (Bedono Hamlet) and observation station 70 (Karangwaru Hamlet Cemetery)</p>

K-fold matching model for crowd behavioral anomaly detection from discontinuous inputs

Image and associated features detect behavioural anomalies in crowd scenes through matching. Existingmethods use continuous data inputs to track crowd behaviour. As a result, the systemmistakenly assumes normal behaviour during data gaps, leading to false positives or missing detections of small abnormalities. The design of precision-focused matching requires simplified yet robust validation for any number of crowd elements. This article presents a K-fold Matching Model (KFMM) with edge-to-edge correlation. This proposed model identifies the region edges of an input crowd image sequentially. The difference in edges due to crowd misbehaviour activities is validated using random and continuous image features in K number of steps. Considering the complexity of multiple K-steps, the confinement is performed using a neural sigmoid function where the K × K edge matching is reduced. This reduction is based on activity classification i.e. minimum and maximum between two consecutive input sequences. Therefore, as the classification increases, the K is halted with the maximum correlation; contrarily for a low number of classifications, the manifold matching is pursued until the maximum is reached. This correlation count is defined using the neural sigmoid with a linear function for the edges. For the maximum features, this proposed model achieves an 11.21 % high matching ratio, 8.15 % high classifications, 9.71 % high precision, 9.61 % less complexity, and 8.9 % less linear error.

Adopting health revenue cycle management best practices among public or private healthcare providers in Saudi Arabia: a pilot study

AbstractThe new era of healthcare providers in Saudi Arabia where all covered beneficiaries will receive valued based medicine with cost effectiveness. Healthcare Transformation in Saudi Arabia: healthcare transformation is aiming to restructure the healthcare providers by Health Sector Transformation Program (HSTP) including revenue cycle management project; which is the core of our study. The literature lacks evidence of scoring systems for health revenue cycle management (HRCM) assessment in Saudi Arabia. We established a gap analysis scoring system to classify healthcare providers’ HRCM readiness. This pilot study aimed to assess the rate of adopting health revenue cycle management best practices among public or private healthcare providers in Saudi Arabia. This pilot study included ten healthcare providers in Saudi Arabia. The providers were assessed during the period from August 2022 to May 2023. The HRCM phases were: contract management, pre-service, time-of-service, and post-service. Contract management and preservice assessment revealed that all of the healthcare providers had missed some or most of their sub-major functions. For time-of-service, all healthcare providers required comprehensive gap analysis and the establishment of policies and procedures. Post-service was assessed in seven healthcare providers but was absent in three. All healthcare providers required training and the activation of reporting protocols and KPIs. Overall, the assessment revealed that one HCP had a low classification, seven had a moderate classification, and two had a high classification. All healthcare providers can be enhanced by training, comprehensive gap analysis, implementing policies and procedures with ongoing monitoring. Additionally, patient satisfaction research and the development of IT/technical solutions.

Pengaruh Model Discovery Learning Dengan Pendekatan STEM Terhadap Hasil Belajar Peserta Didik Kelas VIII SMP Negeri 3 Lubuk Basung

Student learning outcomes are often unsatisfactory due to learning methods that still dominantly use lectures and do not actively involve students. This method tends for making learner passive at studying. For improving this situation and make learning more effective, it so urgent for overhaul the teaching oncoming. One way is for integrate learning models that involve STEM components. The focus through the following discussion the aim is to find out “the effect of the Discovery Learning learning model with a STEM approach on the learning outcomes of grade VIII students at SMPN 3 Lubuk Basung”.This research used a quasi-experimental design by Non-Equivalent Control Group Design. This research population consisted of VIII grade students at SMPN 3 Lubuk Basung for the 2024/2025 academic year. The sample method used was purposive sampling, where one VIII grade was selected being experimental group also the other VIII class being control group. The instruments using this study included multiple choice experiment of 20 questions also a student response questionnaire. Data analysing has been implemented with “normality test, homogeneity test, and non-parametric test, namely Mann-Whitney test”. The survey of learner proves that most of them give positive responses to application of the Discovery Learning model by STEM approach through a ratio 45% of learner agreeing (S) and 52% strongly agreeing (SS). The hypothesis experimental results show significant figures with large 0.000 <0.05, meaning that the application of this model significantly improved student learning outcomes compared to the Discovery Learning model that did not integrate STEM. The N-Gain testing at experimental group reached 0.8594 which is included at high classification. The teaching use with the STEM-based Discovery Learning model was recorded at 97%. This finding shows thats the using Discovery Learning combined by STEM approach found positively and significant effect at learner understanding of introduction to cell material.

The Influence of Work-Life Balance and Workload on Employee Performance at Bank BTN Tasikmalaya Branch Office

This research purposse to determine and analyze the picture of work-life balance, workload and employee performance at Bank BTN Tasikmalaya Branch Office. The influence of work-life balance and workload on employee performance simultaneously at Bank BTN Tasikmalaya Branch Office. The influence of work-life balance on employee performance partially at Bank BTN Tasikmalaya Branch Office and the influence of workload on employee performance partially at Bank BTN Tasikmalaya Branch Office. The method used in this research is a survey with a descriptive approach. With a research sample of 35 collection department employees as respondents. The data used is primary and secondary data. The analytical tool used in this research is multiple linear regression using the SPSS version 26 support application. The test results show that work-life balance has a good classification, workload has a good classification and employee performance has a high classification, simultaneously showing the variables used. in the research, work-life balance and workload have a significant effect on employee performance at Bank BTN Tasikmalaya Branch Office. Partially, there is a significant influence of workload on employee performance, while work-life balance does not have a significant influence on employee performance.

Two-step deep learning models for detection and identification of the manufacturers and types of dental implants on panoramic radiographs.

The purpose of this study is to develop two-step deep learning models that can automatically detect implant regions on panoramic radiographs and identify several types of implants. A total of 1,574 panoramic radiographs containing 3675 implants were included. The implant manufacturers were Kyocera, Dentsply Sirona, Straumann, and Nobel Biocare. Model A was created to detect oral implants and identify the manufacturers using You Only Look Once (YOLO) v7. After preparing the image patches that cropped the implant regions detected by model A, model B was created to identify the implant types per manufacturer using EfficientNet. Model A achieved very high performance, with recall of 1.000, precision of 0.979, and F1 score of 0.989. It also had accuracy, recall, precision, and F1 score of 0.98 or higher for the classification of the manufacturers. Model B had high classification metrics above 0.92, exception for Nobel's class 2 (Parallel). In this study, two-step deep learning models were built to detect implant regions, identify four manufacturers, and identify implant types per manufacturer.

Improving science process skills using the STEM (science, technology, engineering, and mathematics) learning model on environmental preservation business material for fifth grade elementary school students

Society must be able to adapt in various ways due to the rapid progress of science and technology in the era of globalization. One of the components is education, which is fundamental for the progress of a country. Science process skills are skills for developing the perspective needed to build natural science. This research was conducted to see whether science process skills improved after applying the STEM learning model to environmental conservation business materials, in class V students at SDN 1 Ciledug located in the city of West Bandung. This research was carried out by designing a pretest-posttest group and using a science process skills test, based on the indicators that had been designed. Data analysis to determine differences in perceptions regarding improving students' science process skills, before and after implementing the STEM learning model. The results of data analysis show that science process abilities have increased on the topic of environmental conservation efforts with the application of the STEM learning model. The resulting score with the average value of science process skills increased from application of 40.51 to 85.11. This improvement in skills can be seen from the N-Gain result value of 0.75 which is in the high classification. Thus, the research results that have been presented suggest that teachers can use the STEM learning model as an alternative in science subjects, especially to improve science process skills.

Artificial intelligence assisted tomato plant monitoring system – An experimental approach based on universal multi-branch general-purpose convolutional neural network

Real-time monitoring of tomato plants in plant factories is necessary to identify and classify diseases at early stages to prevent possible outbreaks. The proposed DeepD381v4plus network exhibits higher class-wise accuracy, sensitivity, specificity, precision, F1 score and Matthews correlation coefficient scores exceeding 0.96 for multi-varietal tomato leaf diseases. During the reproductive stage, bud formation, flower appearance, bite marks and fruit set also need to be monitored to confirm pollination. The detector DeepDet381v4 – YOLOv4M achieves the highest mean average precision (mAP) (0.90) and lowest mAP (0.68) in the TFl_Blooming class and the lowest mAP (0.68) in the TFl_Transforming class. However, in real-world simulations, DeepDet381v4 – YOLOv4M can detect and count ripe tomatoes at a distance of 40 cm with little to no error. Both networks used for classification and detection–counting tasks have small sizes with high classification and detection efficiency (>27 fps). Overall, the proposed experimental approach will help farmers prevent disease outbreaks, monitor flower shapes that can set fruits at the highest rate, timely detect and count ripened fruits or recognise damaged fruits due to surface cracks or diseases for harvesting at their optimal maturity stage. This will reduce the labour costs, improve cultivation management and ensure excellent quality of the harvested tomatoes.

Precision classification and quantitative analysis of bacteria biomarkers via surface-enhanced Raman spectroscopy and machine learning

The SERS spectra of six bacterial biomarkers, 2,3-DHBA, 2,5-DHBA, Pyocyanin, lipoteichoic acid (LTA), Enterobactin, and β-carotene, of various concentrations, were obtained from silver nanorod array substrates, and the spectral peaks and the corresponding vibrational modes were identified to classify different spectra. The spectral variations in three different concentration regions due to various reasons have imposed a challenge to use classic calibration curve methods to quantify the concentration of biomarkers. Depending on baseline removal strategy, i.e., local or global baseline removal, the calibration curve differed significantly. With the aid of convolutional neural network (CNN), a two-step process was established to classify and quantify biomarker solutions based on SERS spectra: using a specific CNN model, a remarkable differentiation and classification accuracy of 99.99 % for all six biomarkers regardless of the concentration can be achieved. After classification, six regression CNN models were established to predict the concentration of biomarkers, with coefficient of determination R2 > 0.97 and mean absolute error (MAE) < 0.27. The feature of important calculations indicates the high classification and quantification accuracies were due to the intrinsic spectral features in SERS spectra. This study showcases the synergistic potential of SERS and advanced machine learning algorithms and holds significant promise for bacterial infection diagnostics.

The Role Of Science Teachers In Increasing Students' Learning Motivation In The Independent Curriculum Era According To Behavioristics

In the era of the Merdeka Curriculum, which is integrated with technological developments, requires a science teacher to carry out his duties professionally. One part of the teacher's professionalism is understanding learning theory. Behavioristic learning theory, which tends to be teacher-centered, is still often used by teachers even though the curriculum has changed because the application of this theory in learning is still relevant and works well. One of the results of research conducted by Yuliani (2020) entitled “Development of E-Book Learning Media with a Behavioristic Approach to Science Subjects for Class IX Students of SMP 24 Banjarmasin” states that the behavioristic approach shows a positive influence on improving student achievement. The results of analyzing student learning using the N-Gain test are 0.8 (included in the high classification). The results of this study indicate that the behavioristic flow in the era of the independent curriculum is able to increase learning motivation even though the application of behavioristic flow tends to be teacher-centered. The urgency of behavioristic flow in the implementation of learning is to maximize the role of teachers, especially in improving student discipline.

Feasibility and efficacy of minimally invasive limited resection for primary duodenal gastrointestinal stromal tumors: a retrospective cohort study

BackgroundThe primary duodenal gastrointestinal stromal tumor (GIST) is a rare type of gastrointestinal tract tumor. Limited resection (LR) has been increasingly performed for duodenal GIST. However, only a few studies reported minimally invasive limited resection (MI-LR) for primary duodenal GIST.MethodsThe clinical data of 33 patients with primary duodenal GIST from December 2014 to February 2024 were retrospectively analyzed including 23 who received MI-LR and 10 who received laparoscopic or robotic pancreaticoduodenectomy (LPD/RPD).ResultsA total of 33 patients with primary duodenal GIST were enrolled and retrospectively reviewed. Patients received MI-LR exhibited less OT (280 vs. 388.5min, P=0.004), EBL (100 vs. 450ml, P<0.001), and lower morbidity of postoperative complications (52.2% vs. 100%, P=0.013) than LPD/RPD. Patients received LPD/RPD burdened more aggressive tumors with larger size (P=0.047), higher classification (P<0.001), and more mitotic count/50 HPF(P=0.005) compared with patients received MI-LR. The oncological outcomes were similar in MI-LR group and LPD/RPD group.All the patients underwent MI-LR with no conversion, including 12 cases of LLR and 11 cases of RLR. All of the clinicopathological data of the patients were similar in both groups. The median OT was 280(210-480) min and 257(180-450) min, and the median EBL was 100(20-1000) mL and 100(20-200) mL in the LLR and the RLR group separately. The postoperative complications mainly included DGE (LLR 4 cases, 33.4% and RLR 4 cases, 36.4%), intestinal fistula (LLR 2 cases, 16.7%, and RLR 0 case), gastrointestinal hemorrhage (LLR 0 case and RLR 1 case, 9.1%), and intra-abdominal infection (LLR 3 cases, 25.0% and RLR 1 case, 9.1%). The median postoperative length of hospitalization was 19.5(7-46) days in the LLR group and 19(9-38) days in the RLR group. No anastomotic stenosis, local recurrence or distant metastasis was observed during the follow-up period in the two groups.ConclusionsMinimally invasive limited resection is an optional treatment for primary duodenal GIST with satisfactory short-term and long-term oncological outcomes.

SPATIO TEMPORAL ANALYSIS OF URBAN HEAT ISLAND USING LANDSAT 8 OLI/TIRS IMAGERY IN KLATEN DISTRICT IN 2013 – 2021

Klaten Regency is one of the areas that has experienced changes in use over time, such as agricultural land becoming built-up or non-agricultural land. This has an impact on the increase in air temperature which can lead to the occurrence of the phenomenon of the Heat Island or Urban Heat Island. This study aims to determine the spatio-temporal variation of vegetation index value (NDVI), Surface Temperature and in Klaten Regency in 2013 - 2021 and determine the relationship between vegetation index value (NDVI) and UHI in Klaten Regency. The method used in this research is Landsat 8 image data processing such as NDVI, LST and UHI calculation to get the final result. Meanwhile, to get the relationship between NDVI and Surface Temperature related to the UHI phenomenon, simple regression testing was carried out. The results obtained from the processing results show that Klaten Regency is dominated by vegetation cover density with very high and high classifications despite the decrease in area from 2013 to 2021. Surface temperature in Klaten Regency is dominated by surface temperatures within the range of 25 - 28o C spread across the entire region from 2013 to 2021. The results of the linear regression have a coefficient value of r2 where the value shows results with a value of 0.377. The result of the coefficient of determination shows that the amount of influence obtained from the density of vegetation land cover with an average surface temperature of 37.7% where the rest can be influenced by other factors.

An Optimized Recurrent Neural Network for re-modernize food dining bowls and estimating food capacity from images

Maintaining health and preventing chronic diseases requires awareness of the energy and nutrient content of one's diet. Due to the rapid development of computing and Artificial intelligence (AI) technology, nutritional evaluation can be performed using smartphones or wearable food images.Thechallenges of this technology include estimating the visual estimation of the amount of food in a bowl. We describe an innovative approach for determining a bowl's dimensions by sticking a paper ruler throughout the bottom and edges yet photographing the outcome. We used a variety of colored liquids, bowlsand crystallized foods to evaluate the precision of our technique for measuring food volume employing spherical bowls as boxes. Different food-related datasets can have other numbers of images, types of food dining bowls and estimated food capacity from the images represented. Image preprocessing using a wiener filter enhances the estimating of food dining bowl capacity from images. Following preprocessing, a threshold-based approach to image segmentation is used for the resulting data; finally, we used the Gradient Lion Swarm Optimized Discrete Recurrent Neural Network (GLSO-DRNN) model to present an AI-based multi-dish food identification approach. An excellent mAP = 0.96 was found between the results. The suggested model shows great promise as a means of improving dish reporting because of its high classification performance.The GLSO-DRNN performed better toproving its flexibility and efficient discrete parameter optimization. In addition to offering accurate food quantity estimates, it demonstrated significant potential for other uses in the field of smart dining technology. The results highlight aneffective ofthis advanced computational method that is improving food dining bowlsand interactions with components associated with food. In the future, we plan to improve the techniques' performance and incorporate our technology into a working Smartphone based on the cloud environment.

Evaluation of Self-Inflicted versus Non-Self-Inflicted Gunshot Wounds and Associated Injuries Involving the Hand and Upper Extremity.

Orthopedic costs associated with gunshot wounds (GSWs) totaled approximately USD 510 million from 2005 to 2014. Previous studies have identified differences in injuries associated with self-inflicted (SI) GSWs; however, there remains a gap in understanding injury patterns. This study aims to expand upon the current literature and shed light on injury patterns and outcomes associated with SI vs. non-self-inflicted (NSI) GSWs. This is a retrospective cohort study of upper extremity GSWs from January 2012 to December 2022. Data were analyzed using the two-sample t-test, Pearson's chi-squared test, and Fisher's exact test. SI GSWs tended to be high-velocity GSWs and occurred more often in distal locations compared to NSI GSWs (p = 0.0014 and p < 0.0001, respectively). SI GSWs were associated with higher Gustilo-Anderson (GA) and Tscherne classifications (p < 0.0001 and p = 0.0048, respectively) and with a greater frequency of neurovascular damage (p = 0.0048). There was no difference in fracture rate or need for operative intervention between the groups. GA and Tscherne classifications were associated with the need for and type of surgery (p < 0.0001), with a higher classification being associated with more intricate operative intervention; however, GSW velocity was not associated with operative need (p = 0.42). Our findings demonstrate that velocity, wound grading systems, and other factors are associated with the manner in which GSWs to the upper extremity are inflicted and may thus have potential for use in the prediction of injury patterns and planning of trauma management and surgical intervention.

Flexible variable selection in the presence of missing data.

In many applications, it is of interest to identify a parsimonious set of features, or panel, from multiple candidates that achieves a desired level of performance in predicting a response. This task is often complicated in practice by missing data arising from the sampling design or other random mechanisms. Most recent work on variable selection in missing data contexts relies in some part on a finite-dimensional statistical model, e.g., a generalized or penalized linear model. In cases where this model is misspecified, the selected variables may not all be truly scientifically relevant and can result in panels with suboptimal classification performance. To address this limitation, we propose a nonparametric variable selection algorithm combined with multiple imputation to develop flexible panels in the presence of missing-at-random data. We outline strategies based on the proposed algorithm that achieve control of commonly used error rates. Through simulations, we show that our proposal has good operating characteristics and results in panels with higher classification and variable selection performance compared to several existing penalized regression approaches in cases where a generalized linear model is misspecified. Finally, we use the proposed method to develop biomarker panels for separating pancreatic cysts with differing malignancy potential in a setting where complicated missingness in the biomarkers arose due to limited specimen volumes.

Language uncovers visuospatial dysfunction in posterior cortical atrophy: a natural language processing approach.

Posterior Cortical Atrophy (PCA) is a syndrome characterized by a progressive decline in higher-order visuospatial processing, leading to symptoms such as space perception deficit, simultanagnosia, and object perception impairment. While PCA is primarily known for its impact on visuospatial abilities, recent studies have documented language abnormalities in PCA patients. This study aims to delineate the nature and origin of language impairments in PCA, hypothesizing that language deficits reflect the visuospatial processing impairments of the disease. We compared the language samples of 25 patients with PCA with age-matched cognitively normal (CN) individuals across two distinct tasks: a visually-dependent picture description and a visually-independent job description task. We extracted word frequency, word utterance latency, and spatial relational words for this comparison. We then conducted an in-depth analysis of the language used in the picture description task to identify specific linguistic indicators that reflect the visuospatial processing deficits of PCA. Patients with PCA showed significant language deficits in the visually-dependent task, characterized by higher word frequency, prolonged utterance latency, and fewer spatial relational words, but not in the visually-independent task. An in-depth analysis of the picture description task further showed that PCA patients struggled to identify certain visual elements as well as the overall theme of the picture. A predictive model based on these language features distinguished PCA patients from CN individuals with high classification accuracy. The findings indicate that language is a sensitive behavioral construct to detect visuospatial processing abnormalities of PCA. These insights offer theoretical and clinical avenues for understanding and managing PCA, underscoring language as a crucial marker for the visuospatial deficits of this atypical variant of Alzheimer's disease.

Association of glucose metabolism and insulin resistance with feed efficiency and production traits of finishing beef steers.

Increasing nutrient utilization efficiency is an important component of enhancing the sustainability of beef cattle production. The objective of this experiment was to determine the association of glucose metabolism and insulin resistance with dry matter intake (DMI), average daily gain (ADG), gain:feed ratio (G:F), and residual feed intake (RFI). Steers (n = 54; initial body weight = 518 ± 27.0kg) were subjected to an intravenous glucose tolerance test (IVGTT) where glucose was dosed through a jugular catheter and serial blood samples were collected. Three days after the last group's IVGTT, steers began a 63-d DMI and ADG test. Body weight was measured on days 0, 1, 21, 42, 62, and 63, and DMI was measured using an Insentec Roughage Intake Control system (Hokofarm Group, Emmeloord, the Netherlands). To examine relationships between DMI, ADG, G:F, and RFI with IVGTT measurements, Pearson correlations were calculated using Proc Corr of SAS 9.4 (SAS Inst. Inc., Cary, NC). Additionally, cattle were classified based on DMI, ADG, RFI, and G:F, where the medium classification was set as mean ± 0.5 SD, the low classification was < 0.5 SD from the mean, and the high classification was > 0.5 SD from the mean. No associations between DMI and IVGTT parameters were observed, and no differences were detected when classifying cattle as having low, medium, or high DMI. Peak insulin concentration in response to the IVGTT tended to be correlated with ADG (r = 0.28; P = 0.07), indicating cattle with greater ADG tend to have a greater insulin release in response to glucose. Glucose nadir concentrations tended to be positively correlated with ADG (r = 0.26; P = 0.10). Additionally, the glucose nadir was greater in high-ADG steers (P = 0.003). The association of greater glucose nadir with high-ADG could indicate that high-ADG steers do not clear glucose as efficiently as low-ADG steers, potentially indicating increased insulin resistance. Further, RFI was not correlated with IVGTT measurements, but low RFI steers had a greater peak glucose concentration (P = 0.040) and tended to have a greater glucose area under the curve (P = 0.09). G:F was correlated with glucose area under the curve (r = 0.33; P = 0.050), glucose nadir (r = 0.35; P = 0.011), and insulin time to peak (r = 0.39; P = 0.010). These results indicate that glucose metabolism and insulin signaling are associated with growth and efficiency, but the molecular mechanisms that drive these effects need to be elucidated.

The Impact of Covid-19 Pandemic on Mental Workload and Fatigue of Students during Online and Offline Practicum

Learning activities underwent policy changes during the Covid-19 pandemic were carried out online and offline. During the Covid-19 pandemic, there were many ways that universities did to keep learning going, such as implementing online learning and online learning. Practical learning experiences changes in the teaching and learning process system due to university policies. This study aims to measure the level of mental workload and the level of student fatigue during online and offline practicums and to determine the relationship between mental workload and fatigue level. The methods used in this research are the National Aeronautics and Space Administration Task Load Index (NASA-TLX) and Subjective Self Rating Test (SSRT) methods. The results showed that the level of mental workload of students was in the high classification with the NASA-TLX dimensions of effort, mental needs, and performance. The level of student fatigue is in the moderate classification with part of the SSRT, namely the weakening of activities. The results of the correlation test show that there is a relationship between mental workload and student fatigue levels with a positive direction of relationship.

Automated and reliable detection of multi-diseases on chest X-ray images using optimized ensemble transfer learning

Chest radiography is a comparatively low-cost, most probably applied in a medical process that brings crucial information to discover diagnostic conclusions. Chest X-rays (CXR) are used in the prognosis of chest diseases that includes asthma, lung cancer, pneumonia, and COVID-19. Artificial intelligence detects the multi-disease automatically to enhance its proficiency and performance by solving image detection complications with various machine learning and deep learning approaches. Convolutional Neural Network (CNN) has been designed for the advancement of computerized recognition systems. For the classification of medical images, texture, shape, size, and tissue constitution are essential features for detecting diseases. Hence, huge input features are merged with deep CNN validated to increase the effectiveness of CXR analysis. In addition, multi-scale features are visualized in CNN for the detection of variable sizes of thoracic diseases. This paper has implemented a novel multi-disease diagnosis model using chest X-ray images through deep learning approaches, and thus, it helps to minimize the computational cost. The standard CXR images are gathered from the standard datasets. After, the pre-processing is performed for cleaning and contrast the images. Further, the image segmentation is carried out using Optimized DeepLabv3 (ODeepLabv3), where the optimization of parameters in DeepLabv3 is done by a new Mutation Rate-based Lion Algorithm (MR-LA). Then, multi-disease classification is carried out through Optimized Ensemble Transfer Learning (OETL), where the OETL is carried out through VGG16, ResNet, ImageNet, MobileNet, GoogleNet, Inception, and Xception. Here, the parameter optimization of all models is done by the same MR-LA. The proposed model improved the effectiveness and performance in expressions of sensitivity, precision, and specificity factors and get higher classification and detection accuracy. The outcome of the recommended method over other models is identified by the relative analysis conducted.

Application of the Learning Vector Quantization Algorithm for Classification of Students with the Potential to Drop Out

Universities, as providers of academic education services, are required to provide an optimal educational process for students to produce a generation of quality human beings. Student learning success is seen as the success of universities in implementing the higher education process. One of the problems universities face in maintaining the quality of education is student dropout. The high dropout rate in universities can impact accreditation assessments. As a result, it will affect the level of public trust. The number of dropouts in higher education can be minimized from an early age by analyzing the factors that cause student dropouts using data on students who graduated and those who dropped out. This data can be used to determine student dropout patterns by classifying them using the artificial neural network learning vector quantization (LVQ) approach. The data used in this research was 4053, consisting of 3840 graduate student data and 213 dropout student data. This data is considered unbalanced, an unbalanced dataset can cause errors because the model tends to classify the majority class with a high classification and pays less attention to the minority class. So, it is necessary to apply oversampling techniques to overcome this problem. The research results show that the application of the LVQ method to unbalanced data produces an accuracy value of 95.53%, a precision value of 100%, a recall value of 15.02% and an f1-score of 0.26, while the application of the LVQ method to data that has undergone resampling resulting in an accuracy value of 94.66%, a precision value of 92.22%, a recall value of 97.55%, and an f1-score value of 0.95. The LVQ method can be used to classify dropout students with excellent results.