Search results for: predictive analysis

Authors: Surapol Raadnui

Abstract:

This particular work describes an experimental investigation on gear wear in which wear and pitting were intentionally allowed to occur, namely, moisture corrosion pitting, acid-induced corrosion pitting, hard contaminant-related pitting and mechanical induced wear. A back to back spur gear test rig and a grease lubricated worm gear rig were used. The tests samples of wear debris were collected and assessed through the utilization of an optical microscope in order to correlate and compare the debris morphology to pitting and wear degradation of the worn gears. In addition, weight loss from all test gear pairs were assessed with utilization of statistical design of experiment. It can be deduced that wear debris characteristics from both cases exhibited a direct relationship with different pitting and wear modes. Thus, it should be possible to detect and diagnose gear pitting and wear utilization of worn surfaces, generated wear debris and quantitative measurement such as weight loss.

Keywords: predictive maintenance, worm gear, spur gear, wear debris analysis, problem diagnostic

Procedia PDF Downloads 145

Authors: Saurabh Sharma, Yong Sun, Bruce Vernham

Abstract:

Accurate prediction of NOx emission is a continuous challenge in the field of diesel engine-out emission modeling. Performing experiments for each conditions and scenario cost significant amount of money and man hours, therefore model-based development strategy has been implemented in order to solve that issue. NOx formation is highly dependent on the burn gas temperature and the O₂ concentration inside the cylinder. The current empirical models are developed by calibrating the parameters representing the engine operating conditions with respect to the measured NOx. This makes the prediction of purely empirical models limited to the region where it has been calibrated. An alternative solution to that is presented in this paper, which focus on the utilization of in-cylinder combustion parameters to form a predictive semi-empirical NOx model. The result of this work is shown by developing a fast and predictive NOx model by using the physical parameters and empirical correlation. The model is developed based on the steady state data collected at entire operating region of the engine and the predictive combustion model, which is developed in Gamma Technology (GT)-Power by using Direct Injected (DI)-Pulse combustion object. In this approach, temperature in both burned and unburnt zone is considered during the combustion period i.e. from Intake Valve Closing (IVC) to Exhaust Valve Opening (EVO). Also, the oxygen concentration consumed in burnt zone and trapped fuel mass is also considered while developing the reported model.  Several statistical methods are used to construct the model, including individual machine learning methods and ensemble machine learning methods. A detailed validation of the model on multiple diesel engines is reported in this work. Substantial numbers of cases are tested for different engine configurations over a large span of speed and load points. Different sweeps of operating conditions such as Exhaust Gas Recirculation (EGR), injection timing and Variable Valve Timing (VVT) are also considered for the validation. Model shows a very good predictability and robustness at both sea level and altitude condition with different ambient conditions. The various advantages such as high accuracy and robustness at different operating conditions, low computational time and lower number of data points requires for the calibration establishes the platform where the model-based approach can be used for the engine calibration and development process. Moreover, the focus of this work is towards establishing a framework for the future model development for other various targets such as soot, Combustion Noise Level (CNL), NO₂/NOx ratio etc.

Keywords: diesel engine, machine learning, NOₓ emission, semi-empirical

Procedia PDF Downloads 110

Authors: V. Pourmostaghimi, M. Zadshakoyan

Abstract:

Efficiency and productivity of the finish hard turning can be enhanced impressively by utilizing accurate predictive models for cutting tool wear. However, the ability of genetic programming in presenting an accurate analytical model is a notable characteristic which makes it more applicable than other predictive modeling methods. In this paper, the genetic equation for modeling of tool flank wear is developed with the use of the experimentally measured flank wear values and genetic programming during finish turning of hardened AISI D2. Series of tests were conducted over a range of cutting parameters and the values of tool flank wear were measured. On the basis of obtained results, genetic model presenting connection between cutting parameters and tool flank wear were extracted. The accuracy of the genetically obtained model was assessed by using two statistical measures, which were root mean square error (RMSE) and coefficient of determination (R²). Evaluation results revealed that presented genetic model predicted flank wear over the study area accurately (R² = 0.9902 and RMSE = 0.0102). These results allow concluding that the proposed genetic equation corresponds well with experimental data and can be implemented in real industrial applications.

Keywords: cutting parameters, flank wear, genetic programming, hard turning

Procedia PDF Downloads 174

Authors: Jie Lin, Yiwen Pan, Li Zhang, Zhangyong Xia

Abstract:

Atherosclerosis is a chronic inflammatory disease characterized by the accumulation of lipids, immune cells, and extracellular matrix in the arterial walls. This pathological process can lead to the formation of plaques that can obstruct blood flow and trigger various cardiovascular diseases such as heart attack and stroke. The underlying molecular mechanisms still remain unclear, although many studies revealed the dysfunction of endothelial cells, recruitment and activation of monocytes and macrophages, and the production of pro-inflammatory cytokines and chemokines in atherosclerosis. This study aimed to identify hub genes involved in the progression of atherosclerosis and to analyze their biological function in silico, thereby enhancing our understanding of the disease’s molecular mechanisms. Through the analysis of microarray data, we examined the gene expression in media and neo-intima from plaques, as well as distant macroscopically intact tissue, across a cohort of 32 hypertensive patients. Initially, 112 differentially expressed genes (DEGs) were identified. Subsequent immune infiltration analysis indicated a predominant presence of 27 immune cell types in the atherosclerosis group, particularly noting an increase in monocytes and macrophages. In the Weighted gene co-expression network analysis (WGCNA), 10 modules with a minimum of 30 genes were defined as key modules, with blue, dark, Oliver green and sky-blue modules being the most significant. These modules corresponded respectively to monocyte, activated B cell, and activated CD4 T cell gene patterns, revealing a strong morphological-genetic correlation. From these three gene patterns (modules morphology), a total of 2509 key genes (Gene Significance >0.2, module membership>0.8) were extracted. Six hub genes (CD36, DPP4, HMOX1, PLA2G7, PLN2, and ACADL) were then identified by intersecting 2509 key genes, 102 DEGs with lipid-related genes from the Genecard database. The bio-functional analysis of six hub genes was estimated by a robust classifier with an area under the curve (AUC) of 0.873 in the ROC plot, indicating excellent efficacy in differentiating between the disease and control group. Moreover, PCA visualization demonstrated clear separation between the groups based on these six hub genes, suggesting their potential utility as classification features in predictive models. Protein-protein interaction (PPI) analysis highlighted DPP4 as the most interconnected gene. Within the constructed key gene-drug network, 462 drugs were predicted, with ursodeoxycholic acid (UDCA) being identified as a potential therapeutic agent for modulating DPP4 expression. In summary, our study identified critical hub genes implicated in the progression of atherosclerosis through comprehensive bioinformatic analyses. These findings not only advance our understanding of the disease but also pave the way for applying similar analytical frameworks and predictive models to other diseases, thereby broadening the potential for clinical applications and therapeutic discoveries.

Keywords: atherosclerosis, hub genes, drug prediction, bioinformatics

Procedia PDF Downloads 60

Authors: Martín Pratto Burgos

Abstract:

The Engineering School of the University of the Republic in Uruguay offers an Introductory Mathematical Course from the second semester of 2019. This course has been designed to assist students in preparing themselves for math courses that are essential for Engineering Degrees, namely Math1, Math2, and Math3 in this research. The research proposes to build a model that can accurately predict the student's activity and academic progress based on their performance in the three essential Mathematical courses. Additionally, there is a need for a model that can forecast the incidence of the Introductory Mathematical Course in the three essential courses approval during the first academic year. The techniques used are Principal Component Analysis and predictive modelling using the Generalised Linear Model. The dataset includes information from 5135 engineering students and 12 different characteristics based on activity and course performance. Two models are created for a type of data that follows a binomial distribution using the R programming language. Model 1 is based on a variable's p-value being less than 0.05, and Model 2 uses the stepAIC function to remove variables and get the lowest AIC score. After using Principal Component Analysis, the main components represented in the y-axis are the approval of the Introductory Mathematical Course, and the x-axis is the approval of Math1 and Math2 courses as well as student activity three years after taking the Introductory Mathematical Course. Model 2, which considered student’s activity, performed the best with an AUC of 0.81 and an accuracy of 84%. According to Model 2, the student's engagement in school activities will continue for three years after the approval of the Introductory Mathematical Course. This is because they have successfully completed the Math1 and Math2 courses. Passing the Math3 course does not have any effect on the student’s activity. Concerning academic progress, the best fit is Model 1. It has an AUC of 0.56 and an accuracy rate of 91%. The model says that if the student passes the three first-year courses, they will progress according to the timeline set by the curriculum. Both models show that the Introductory Mathematical Course does not directly affect the student’s activity and academic progress. The best model to explain the impact of the Introductory Mathematical Course on the three first-year courses was Model 1. It has an AUC of 0.76 and 98% accuracy. The model shows that if students pass the Introductory Mathematical Course, it will help them to pass Math1 and Math2 courses without affecting their performance on the Math3 course. Matching the three predictive models, if students pass Math1 and Math2 courses, they will stay active for three years after taking the Introductory Mathematical Course, and also, they will continue following the recommended engineering curriculum. Additionally, the Introductory Mathematical Course helps students to pass Math1 and Math2 when they start Engineering School. Models obtained in the research don't consider the time students took to pass the three Math courses, but they can successfully assess courses in the university curriculum.

Keywords: machine-learning, engineering, university, education, computational models

Procedia PDF Downloads 87

Authors: Rachel C. F. Sun

Abstract:

This study examined the predictive effects of moral competence, prosocial norms and positive behavior recognition on school misbehavior among Chinese junior secondary school students. Results of multiple regression analysis showed that students were more likely to misbehave in school when they had lower levels of moral competence and prosocial norms, and when they perceived their positive behavior being less likely recognized. Practical implications were discussed on how to guide students to make the right choices to behave appropriately in school. Implications for future research were also discussed.

Keywords: moral competence, positive behavior recognition, prosocial norms, school misbehavior

Procedia PDF Downloads 378

Authors: Zakarya Kourdi, Mohammed Khaouani, Benyounes Bouazza, Ahlam Guen-Bouazza, Amine Boursali

Abstract:

In this paper, we have evaluated the thermal effect for high electron mobility transistors (HEMTs) heterostructure InAlN/GaN with a gate length 30nm high-performance. It also shows the analysis and simulated these devices, and how can be used in different application. The simulator Tcad-Silvaco software has used for predictive results good for the DC, AC and RF characteristic, Devices offered max drain current 0.67A; transconductance is 720 mS/mm the unilateral power gain of 180 dB. A cutoff frequency of 385 GHz, and max frequency 810 GHz These results confirm the feasibility of using HEMTs with InAlN/GaN in high power amplifiers, as well as thermal places.

Keywords: HEMT, Thermal Effect, Silvaco, InAlN/GaN

Procedia PDF Downloads 463

Authors: Sharif Muddsair, Eisels Martin, Giesbrecht Eugenie

Abstract:

With development of society increase the demand for the movement of people also increases gradually. The various modes of the transport in different extent which expat impacts, which depends on mainly technical-operating conditions. The up-to-date telematics systems provide the transport industry a revolutionary. Appropriate use of these systems can help to substantially improve the efficiency. Vehicle monitoring and fleet tracking are among services used for improving efficiency and effectiveness of utility vehicle. There are many telematics systems which may contribute to eco-driving. Generally, they can be grouped according to their role in driving cycle. • Before driving - eco-route selection, • While driving – Advanced driver assistance, • After driving – remote analysis. Our point of interest is regulated in third point [after driving – remote analysis]. TS [Telematics-system] make it possible to record driving patterns in real time and analysis the data later on, So that driver- classification-specific hints [fast driver, slow driver, aggressive driver…)] are given to imitate eco-friendly driving style. Together with growing number of vehicle and development of information technology, telematics become an ‘active’ research subject in IT and the car industry. Telematics has gone a long way from providing navigation solution/assisting the driver to become an integral part of the vehicle. Today’s telematics ensure safety, comfort and become convenience of the driver.

Keywords: internet of things, iot, connected vehicle, cv, ts, telematics services, ml, machine learning

Procedia PDF Downloads 298

Authors: Mohd Sabri Minhat, Nurul Adilla Mohd Subha

Abstract:

The PUSPATI TRIGA Reactor (RTP), Malaysia reached its first criticality on June 28, 1982, with power capacity 1MW thermal. The Feedback Control Algorithm (FCA) which is conventional Proportional-Integral (PI) controller, was used for present power control method to control fission process in RTP. It is important to ensure the core power always stable and follows load tracking within acceptable steady-state error and minimum settling time to reach steady-state power. At this time, the system could be considered not well-posed with power tracking performance. However, there is still potential to improve current performance by developing next generation of a novel design nuclear core power control. In this paper, the dual-mode predictions which are proposed in modelling Optimal Model Predictive Control (OMPC), is presented in a state-space model to control the core power. The model for core power control was based on mathematical models of the reactor core, OMPC, and control rods selection algorithm. The mathematical models of the reactor core were based on neutronic models, thermal hydraulic models, and reactivity models. The dual-mode prediction in OMPC for transient and terminal modes was based on the implementation of a Linear Quadratic Regulator (LQR) in designing the core power control. The combination of dual-mode prediction and Lyapunov which deal with summations in cost function over an infinite horizon is intended to eliminate some of the fundamental weaknesses related to MPC. This paper shows the behaviour of OMPC to deal with tracking, regulation problem, disturbance rejection and caters for parameter uncertainty. The comparison of both tracking and regulating performance is analysed between the conventional controller and OMPC by numerical simulations. In conclusion, the proposed OMPC has shown significant performance in load tracking and regulating core power for nuclear reactor with guarantee stabilising in the closed-loop.

Keywords: core power control, dual-mode prediction, load tracking, optimal model predictive control

Procedia PDF Downloads 155

Authors: A. Saimen, E. Armstrong, C. Manitshana

Abstract:

Intimate partner violence (IPV) has been recognised as a global human rights violation. It is universally under diagnosed and the institution of timeous multi-faceted interventions has been noted to benefit IPV victims. Currently, the concept of using a screening tool to detect IPV has not been widely explored in a primary healthcare setting in South Africa, and it was for this reason that this study has been undertaken. A systematic random sampling of 1 in 8 women over a period of 3 months was conducted prospectively at the OPD of a Level 1 Hospital. Participants were asked about their experience of IPV during the past 12 months. The WAST-short, a two-question tool, was used to screen patients for IPV. To verify the result of the screening, women were also asked the remaining questions from the WAST. Data was collected from 400 participants, with a response rate of 99.3%. The prevalence of IPV in the sample was 32%. The WAST-short was shown to have the following operating characteristics: sensitivity 45.2%, specificity 98%,positive predictive value 98%, negative predictive value 79%. The WAST-short lacks sufficient sensitivity and therefore is not an ideal screening tool for this setting. Improvement in the sensitivity of the WAST-short in this setting may be achieved by lowering the threshold for a positive result for IPV screening, and modification of the screening questions to better reflect IPV as understood by the local population.

Keywords: domestic violence, intimate partner violence, screening, screening tools

Procedia PDF Downloads 301

Authors: Muhammad Waleed Asfandyar, Akhtar Ahmed, Syed Adib-ul-Hasan Rizvi

Abstract:

Objective: To evaluate the ability of serum concentration of prostate specific antigen between two cutting points considering it as a predictor of skeletal metastasis on bone scintigraphy in men with prostate cancer. Settings: This study was carried out in department of Nuclear Medicine at Sindh Institute of Urology and Transplantation (SIUT) Karachi, Pakistan. Materials and Method: From August 2013 to November 2013, forty two (42) consecutive patients with prostate cancer who underwent technetium-99m methylene diphosphonate (Tc-99mMDP) whole body bone scintigraphy were prospectively analyzed. The information was collected from the scintigraphic database at a Nuclear medicine department Sindh institute of urology and transplantation Karachi Pakistan. Patients who did not have a serum PSA concentration available within 1 month before or after the time of performing the Tc-99m MDP whole body bone scintigraphy were excluded from this study. A whole body bone scintigraphy scan (from the toes to top of the head) was performed using a whole-body Moving gamma camera technique (anterior and posterior) 2–4 hours after intravenous injection of 20 mCi of Tc-99m MDP. In addition, all patients necessarily have a pathological report available. Bony metastases were determined from the bone scan studies and no further correlation with histopathology or other imaging modalities were performed. To preserve patient confidentiality, direct patient identifiers were not collected. In all the patients, Prostate specific antigen values and skeletal scintigraphy were evaluated. Results: The mean age, mean PSA, and incidence of bone metastasis on bone scintigraphy were 68.35 years, 370.51 ng/mL and 19/42 (45.23%) respectively. According to PSA levels, patients were divided into 5 groups < 10ng/mL (10/42), 10-20 ng/mL (5/42), 20-50 ng/mL (2/42), 50-100 (3/42), 100- 500ng/mL (3/42) and more than 500ng/mL (0/42) presenting negative bone scan. The incidence of positive bone scan (%) for bone metastasis for each group were O1 patient (5.26%), 0%, 03 patients (15.78%), 01 patient (5.26%), 04 patients (21.05%), and 10 patients (52.63%) respectively. From the 42 patients 19 (45.23%) presented positive scintigraphic examination for the presence of bone metastasis. 1 patient presented bone metastasis on bone scintigraphy having PSA level less than 10ng/mL, and in only 1 patient (5.26%) with bone metastasis PSA concentration was less than 20 ng/mL. therefore, when the cutting point adopted for PSA serum concentration was 10ng/mL, a negative predictive value for bone metastasis was 95% with sensitivity rates 94.74% and the positive predictive value and specificities of the method were 56.53% and 43.48% respectively. When the cutting point of PSA serum concentration was 20ng/mL the observed results for Positive predictive value and specificity were (78.27% and 65.22% respectively) whereas negative predictive value and sensitivity stood (100% and 95%) respectively. Conclusion: Results of our study allow us to conclude that serum PSA concentration of higher than 20ng/mL was the most accurate cutting point than a serum concentration of PSA higher than 10ng/mL to predict metastasis in radionuclide bone scintigraphy. In this way, unnecessary cost can be avoided, since a considerable part of prostate adenocarcinomas present low serum PSA levels less than 20 ng/mL and for these cases radionuclide bone scintigraphy could be unnecessary.

Keywords: bone scan, cut off value, prostate specific antigen value, scintigraphy

Procedia PDF Downloads 312

Authors: Sagri Sharma

Abstract:

Analysis of diseases integrating multi-factors increases the complexity of the problem and therefore, development of frameworks for the analysis of diseases is an issue that is currently a topic of intense research. Due to the inter-dependence of the various parameters, the use of traditional methodologies has not been very effective. Consequently, newer methodologies are being sought to deal with the problem. Supervised Learning Algorithms are commonly used for performing the prediction on previously unseen data. These algorithms are commonly used for applications in fields ranging from image analysis to protein structure and function prediction and they get trained using a known dataset to come up with a predictor model that generates reasonable predictions for the response to new data. Gene expression profiles generated by DNA analysis experiments can be quite complex since these experiments can involve hypotheses involving entire genomes. The application of well-known machine learning algorithm - Support Vector Machine - to analyze the expression levels of thousands of genes simultaneously in a timely, automated and cost effective way is thus used. The objectives to undertake the presented work are development of a methodology to identify genes relevant to Hepatocellular Carcinoma (HCC) from gene expression dataset utilizing supervised learning algorithms and statistical evaluations along with development of a predictive framework that can perform classification tasks on new, unseen data.

Keywords: artificial intelligence, biomarker, gene expression datasets, hepatocellular carcinoma, machine learning, supervised learning algorithms, support vector machine

Procedia PDF Downloads 423

Authors: Mu-Yen Chen, Min-Hsuan Fan, Chia-Chen Chen, Siang-Yu Jhong

Abstract:

In recent years, real estate prediction or valuation has been a topic of discussion in many developed countries. Improper hype created by investors leads to fluctuating prices of real estate, affecting many consumers to purchase their own homes. Therefore, scholars from various countries have conducted research in real estate valuation and prediction. With the back-propagation neural network that has been popular in recent years and the orthogonal array in the Taguchi method, this study aimed to find the optimal parameter combination at different levels of orthogonal array after the system presented different parameter combinations, so that the artificial neural network obtained the most accurate results. The experimental results also demonstrated that the method presented in the study had a better result than traditional machine learning. Finally, it also showed that the model proposed in this study had the optimal predictive effect, and could significantly reduce the cost of time in simulation operation. The best predictive results could be found with a fewer number of experiments more efficiently. Thus users could predict a real estate transaction price that is not far from the current actual prices.

Keywords: artificial neural network, Taguchi method, real estate valuation model, investors

Procedia PDF Downloads 482

Authors: Rao Fu, Jie Chen, Shutian Zeng, Yiping Zhuang, Agus Sudjianto

Abstract:

Generative Adversarial Net (GAN) has proved to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions include both categorical and continuous variables with different auxiliary information. Our simulation studies show that CGAN has the capability to learn different types of normal and heavy-tailed distributions, as well as dependent structures of different time series. It also has the capability to generate conditional predictive distributions consistent with training data distributions. We also provide an in-depth discussion on the rationale behind GAN and the neural networks as hierarchical splines to establish a clear connection with existing statistical methods of distribution generation. In practice, CGAN has a wide range of applications in market risk and counterparty risk analysis: it can be applied to learn historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES), and it can also predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate that CGAN can outperform Historical Simulation (HS), a popular method in market risk analysis to calculate VaR. CGAN can also be applied in economic time series modeling and forecasting. In this regard, we have included an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN at the end of the paper.

Keywords: conditional generative adversarial net, market and credit risk management, neural network, time series

Procedia PDF Downloads 138

Authors: Taoufik Benchelha, Toufik Remmal, Rachid El Hamdouni, Hamou Mansouri, Houssein Ejjaouani, Halima Jounaid, Said Benchelha

Abstract:

During the earthworks for the construction of Marrakech-Agadir highway in southern Morocco, which crosses mountainous areas of the High Western Atlas, the main problem faced is the stability of the slopes. Indeed, the use of explosives as a means of excavation associated with the geological structure of the terrain encountered can trigger major ruptures and cause damage which depends on the intrinsic characteristics of the rock mass. The study consists of a geological and geotechnical analysis of several unstable zones located along the route, mobilizing millions of cubic meters of rock, with deduction of the parameters influencing slope stability. From this analysis, a predictive model for rock mass stability is carried out, based on a statistic method of logistic regression, in order to predict the geomechanical behavior of the rock slopes constrained by earthworks.

Keywords: explosive, logistic regression, rock mass, slope stability

Procedia PDF Downloads 369

Authors: Christian Neunzig, Simon Fahle, Jürgen Schulz, Matthias Möller, Bernd Kuhlenkötter

Abstract:

Manufacturing companies are facing global competition and enormous cost pressure. The use of machine learning applications can help reduce production costs and create added value. Predictive quality enables the securing of product quality through data-supported predictions using machine learning models as a basis for decisions on test results. Furthermore, machine learning methods are able to process large amounts of data, deal with unfavourable row-column ratios and detect dependencies between the covariates and the given target as well as assess the multidimensional influence of all input variables on the target. Real production data are often subject to highly fluctuating boundary conditions and unbalanced data sets. Changes in production data manifest themselves in trends, systematic shifts, and seasonal effects. Thus, Machine learning applications require intensive pre-processing and feature selection. Data preprocessing includes rule-based data cleaning, the application of dimensionality reduction techniques, and the identification of comparable data subsets. Within the used real data set of Bosch hydraulic valves, the comparability of the same production conditions in the production of hydraulic valves within certain time periods can be identified by applying the concept drift method. Furthermore, a classification model is developed to evaluate the feature importance in different subsets within the identified time periods. By selecting comparable and stable features, the number of features used can be significantly reduced without a strong decrease in predictive power. The use of cross-process production data along the value chain of hydraulic valves is a promising approach to predict the quality characteristics of workpieces. In this research, the ada boosting classifier is used to predict the leakage of hydraulic valves based on geometric gauge blocks from machining, mating data from the assembly, and hydraulic measurement data from end-of-line testing. In addition, the most suitable methods are selected and accurate quality predictions are achieved.

Keywords: classification, achine learning, predictive quality, feature selection

Procedia PDF Downloads 157

Authors: Wangxu Xia

Abstract:

BACKGROUND BI-RADS 4 lesions raise the possibility of malignancy that warrant further clinical and radiologic work-up. This study aimed to evaluate the predictive performance of diffusion-weighted imaging(DWI) and microcalcifications on mammography for predicting malignancy of BI-RADS 4 lesions. In addition, the predictive performance of DWI combined with microcalcifications was alsocompared with the Kaiser score. METHODS During January 2021 and June 2023, 144 patients with 178 BI-RADS 4 lesions underwent conventional MRI, DWI, and mammography were included. The lesions were dichotomized intobenign or malignant according to the pathological results from core needle biopsy or surgical mastectomy. DWI was performed with a b value of 0 and 800s/mm2 and analyzed using theapparent diffusion coefficient, and a Kaiser score > 4 was considered to suggest malignancy. Thediagnostic performances for various diagnostic tests were evaluated with the receiver-operatingcharacteristic (ROC) curve. RESULTS The area under the curve (AUC) for DWI was significantly higher than that of the of mammography (0.86 vs 0.71, P<0.001), but was comparable with that of the Kaiser score (0.86 vs 0.84, P=0.58). However, the AUC for DWI combined with mammography was significantly highthan that of the Kaiser score (0.93 vs 0.84, P=0.007). The sensitivity for discriminating malignant from benign BI-RADS 4 lesions was highest at 89% for Kaiser score, but the highest specificity of 83% can be achieved with DWI combined with mammography. CONCLUSION DWI combined with microcalcifications on mammography could discriminate malignant BI-RADS4 lesions from benign ones with a high AUC and specificity. However, Kaiser score had a better sensitivity for discrimination.

Keywords: MRI, DWI, mammography, breast disease

Procedia PDF Downloads 53

Authors: Federico Herrera, Valle Gomez García de Soria, Itxaso Portero Sainz, Carlos Fernández Arandojo, Mercedes Royg, Ana Marcos Jimenez, Anna Kreutzman, Cecilia MuñozCalleja

Abstract:

Introduction: Graft-versus-host disease (GVHD) still remains the major complication associated with allogeneic stem cell transplantation (SCT). The pathogenesis involves migration of donor naïve T-cells into recipient secondary lymphoid organs. Two molecules are important in this process: CD62L and CCR7, which are characteristically expressed in naïve/central memory T-cells. With this background, we aimed to study the influence of CCR7 and CD62L on donor lymphocytes in the development and severity of GVHD. Material and methods: This single center study included 98 donor-recipient pairs. Samples were collected prospectively from the apheresis product and phenotyped by flow cytometry. CCR7 and CD62L expression in CD4+ and CD8+ T-cells were compared between patients who developed acute (n=40) or chronic GVHD (n=33) and those who did not (n=38). Results: The patients who developed acute GVHD were transplanted with a higher percentage of CCR7+CD4+ T-cells (p = 0.05) compared to the no GVHD group. These results were confirmed when these patients were divided in degrees according to the severity of the disease; the more severe disease, the higher percentage of CCR7+CD4+ T-cells. Conversely, chronic GVHD patients received a higher percentage of CCR7+CD8+ T-cells (p=0.02) in comparison to those who did not develop the complication. These data were also confirmed when patients were subdivided in degrees of the disease severity. A multivariable analysis confirmed that percentage of CCR7+CD4+ T-cells is a predictive factor of acute GVHD whereas the percentage of CCR7+CD8+ T-cells is a predictive factor of chronic GVHD. In vitro functional assays (migration and activation assays) supported the idea of CCR7+ T-cells were involved in the development of GVHD. As low levels of CD62L expression were detected in all apheresis products, we tested the hypothesis that CD62L was shed during apheresis procedure. Comparing CD62L surface levels in T-cells from the same donor immediately before collecting the apheresis product, and the final apheresis product we found that this process down-regulated CD62L in both CD4+ and CD8+ T cells (p=0.008). Interestingly, when CD62L levels were analysed in days 30 or 60 after engraftment, they recovered to baseline (p=0.008). However, to investigate the relation between CD62L expression and the development of GVHD in the recipient samples after the engraftment, no differences were observed comparing patients with GVHD to those who did not develop the disease. Discussion: Our prospective study indicates that the CCR7+ T-cells from the donor, which include naïve and central memory T-cells, contain the alloreactive cells with a high ability to mediate GVHD (in the case of both migration and activation). Therefore we suggest that the proportion and functional properties of CCR7+CD4+ and CCR7+CD8+ T-cells in the apheresis could act as a predictive biomarker to both acute and chronic GVHD respectively. Importantly, our study precludes that CD62L is lost in the apheresis and therefore it is not a reliable biomarker for the development of GVHD.

Keywords: CCR7, CD62L, GVHD, SCT

Procedia PDF Downloads 284

Authors: Levi Corallo

Abstract:

Typed language made by humans on computers and cell phones has made a significant distinction from previous modes of written language exchanges. While acronyms remain one of the most predominant markings of typed language, another and perhaps more recent revolution in the way humans communicate has been with the use of symbols or glyphs, primarily Emojis—globally introduced on the iPhone keyboard by Apple in 2008. This paper seeks to analyze the use of symbols in typed communication from both a linguistic and machine learning perspective. The Unicode system will be explored and methods of encoding will be juxtaposed with the current machine and human perception. Topics in how typed symbol usage exists in conversation will be explored as well as topics across current research methods dealing with Emojis like sentiment analysis, predictive text models, and so on. This study proposes that sequential analysis is a significant feature for analyzing unicode characters in a corpus with machine learning. Current models that are trying to learn or translate the meaning of Emojis should be starting to learn using bi- and tri-grams of Emoji, as well as observing the relationship between combinations of different Emoji in tandem. The sociolinguistics of an entire new vernacular of language referred to here as ‘typed language’ will also be delineated across my analysis with unicode glyphs from both a semantic and technical perspective.

Keywords: unicode, text symbols, emojis, glyphs, communication

Procedia PDF Downloads 190

Authors: Moti Zwilling, Srečko Natek

Abstract:

This paper aims to present an innovative predictive analytics analysis in order to find the best combination between two consumers who strive to find their partner or in internet sites. The methodology shown in this paper is based on analysis of consumer preferences and involves data mining and machine learning search techniques. The study is composed of two parts: The first part examines by means of descriptive statistics the correlations between a set of parameters that are taken between man and women where they intent to meet each other through the social media, usually the internet. In this part several hypotheses were examined and statistical analysis were taken place. Results show that there is a strong correlation between the affiliated attributes of man and woman as long as concerned to how they present themselves in a social media such as "Facebook". One interesting issue is the strong desire to develop a serious relationship between most of the respondents. In the second part, the authors used common data mining algorithms to search and classify the most important and effective attributes that affect the response rate of the other side. Results exhibit that personal presentation and education background are found as most affective to achieve a positive attitude to one's profile from the other mate.

Keywords: dating sites, social networks, machine learning, decision trees, data mining

Procedia PDF Downloads 289

Authors: Andrea L. Arnett, Ann T. Packard, Yolanda I. Garces, Kenneth W. Merrell

Abstract:

Objective: Pathologic response following neoadjuvant chemoradiation (CRT) has been associated with improved overall survival (OS). Conflicting results have been reported regarding the pathologic predictive value of positron emission tomography (PET) response in patients with stage III lung cancer. The aim of this study was to evaluate the correlation between post-treatment PET response and pathologic response utilizing novel FDG-PET parameters. Methods: This retrospective study included patients with non-metastatic, stage IIIA (N2) NSCLC cancer treated with CRT followed by resection. All patients underwent PET prior to and after neoadjuvant CRT. Univariate analysis was utilized to assess correlations between PET response, nodal clearance, pCR, and near-complete pathologic response (defined as the microscopic residual disease or less). Maximal standard uptake value (SUV), standard uptake ratio (SUR) [normalized independently to the liver (SUR-L) and blood pool (SUR-BP)], metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were measured pre- and post-chemoradiation. Results: A total of 44 patients were included for review. Median age was 61.9 years, and median follow-up was 2.6 years. Histologic subtypes included adenocarcinoma (72.2%) and squamous cell carcinoma (22.7%), and the majority of patients had the T2 disease (59.1%). The rate of pCR and near-complete pathologic response within the primary lesion was 28.9% and 44.4%, respectively. The average reduction in SUVmₐₓ was 9.2 units (range -1.9-32.8), and the majority of patients demonstrated some degree of favorable treatment response. SUR-BP and SUR-L showed a mean reduction of 4.7 units (range -0.1-17.3) and 3.5 units (range –1.7-12.6), respectively. Variation in PET response was not significantly associated with histologic subtype, concurrent chemotherapy type, stage, or radiation dose. No significant correlation was found between pathologic response and absolute change in MTV or TLG. Reduction in SUVmₐₓ and SUR were associated with increased rate of pathologic response (p ≤ 0.02). This correlation was not impacted by normalization of SUR to liver versus mediastinal blood pool. A threshold of > 75% decrease in SUR-L correlated with near-complete response, with a sensitivity of 57.9% and specificity of 85.7%, as well as positive and negative predictive values of 78.6% and 69.2%, respectively (diagnostic odds ratio [DOR]: 5.6, p=0.02). A threshold of >50% decrease in SUR was also significantly associated pathologic response (DOR 12.9, p=0.2), but specificity was substantially lower when utilizing this threshold value. No significant association was found between nodal PET parameters and pathologic nodal clearance. Conclusions: Our results suggest that treatment response to neoadjuvant therapy as assessed on PET imaging can be a predictor of pathologic response when evaluated via SUV and SUR. SUR parameters were associated with higher diagnostic odds ratios, suggesting improved predictive utility compared to SUVmₐₓ. MTV and TLG did not prove to be significant predictors of pathologic response but may warrant further investigation in a larger cohort of patients.

Keywords: lung cancer, positron emission tomography (PET), standard uptake ratio (SUR), standard uptake value (SUV)

Procedia PDF Downloads 226

Authors: Regina Chua

Abstract:

To reduce the costs of re-fielding a large survey, a Random Forest classifier was applied to measure the accuracy of classifying individuals into their assigned segments with the fewest possible questions. Given a long survey, one needed to determine the most predictive ten or fewer questions that would accurately assign new individuals to custom segments. Furthermore, the solution needed to be quick in its classification and usable in non-Python environments. In this paper, a supervised Random Forest classifier was modeled on a dataset with 7,000 individuals, 60 questions, and 254 features. The Random Forest consisted of an iterative collection of individual decision trees that result in a predicted segment with robust precision and recall scores compared to a single tree. A random 70-30 stratified sampling for training the algorithm was used, and accuracy trade-offs at different depths for each segment were identified. Ultimately, the Random Forest classifier performed at 87% accuracy at a depth of 10 with 20 instead of 254 features and 10 instead of 60 questions. With an acceptable accuracy in prioritizing feature selection, new tools were developed for non-Python environments: a worksheet with a formulaic version of the algorithm and an embedded function to predict the segment of an individual in real-time. Random Forest was determined to be an optimal classification model by its feature selection, performance, processing speed, and flexible application in other environments.

Keywords: machine learning, supervised learning, data science, random forest, classification, prediction, predictive modeling

Procedia PDF Downloads 88

Authors: Nicholas Bayfield, Liam Bibo, Charley Budgeon, Robert Larbalestier, Tom Briffa

Abstract:

Introduction: Stress hyperglycaemia is common following cardiac surgery. Its optimal management is uncertain and may differ by diabetic status. This study assesses the in-hospital glycaemic management of cardiac surgery patients and associated postoperative outcomes. Methods: A retrospective cohort analysis of all patients undergoing cardiac surgery at Fiona Stanley Hospital from February 2015 to May 2019 was undertaken. Management and outcomes of hyperglycaemia following cardiac surgery were assessed. Follow-up was assessed to 1 year postoperatively. Multivariate regression modelling was utilised. Results: 1050 non-diabetic patients and 689 diabetic patients were included. In the non-diabetic cohort, patients with mild (peak blood sugar level [BSL] < 14.3), transient stress hyperglycaemia managed without insulin were not at an increased risk of wound-related morbidity (P=0.899) or mortality at 1 year (P=0.483). Insulin management was associated with wound-related readmission to hospital (P=0.004) and superficial sternal wound infection (P=0.047). Prolonged or severe stress hyperglycaemia was predictive of hospital re-admission (P=0.050) but not morbidity or mortality (P=0.546). Diabetes mellitus was an independent risk factor 1-year mortality (OR; 1.972 [1.041–3.736], P=0.037), graft harvest site wound infection (OR; 1.810 [1.134–2.889], P=0.013) and wound-related readmission (OR; 1.866 [1.076–3.236], P=0.026). In diabetics, postoperative peak BSL > 13.9mmol/L was predictive of graft harvest site infections (OR; 3.528 [1.724-7.217], P=0.001) and wound-related readmission OR; 3.462 [1.540-7.783], P=0.003) regardless of modality of management. A peak BSL of 10.0-13.9 did not increase the risk of morbidity/mortality compared to a peak BSL of < 10.0 (P=0.557). Diabetics with a peak BSL of 13.9 or less did not have significantly increased morbidity/mortality outcomes compared to non-diabetics (P=0.418). Conclusion: In non-diabetic patients, transient mild stress hyperglycaemia following cardiac surgery does not uniformly require treatment. In diabetic patients, postoperative hyperglycaemia with peak BSL exceeding 13.9mmol/L was associated with wound-related morbidity and hospital readmission following cardiac surgery.

Keywords: cardiac surgery, pulmonary embolism, pulmonary embolectomy, cardiopulmonary bypass

Procedia PDF Downloads 157

Authors: Shahrzad Zolfagharnassab, Abdul Rashid Mohamed Shariff, Siti Khairunniza Bejo

Abstract:

Conventional methods for nutrient soil mapping are based on laboratory tests of samples that are obtained from surveys. The time and cost involved in gathering and analyzing soil samples are the reasons that researchers use Predictive Soil Mapping (PSM). PSM can be defined as the development of a numerical or statistical model of the relationship among environmental variables and soil properties, which is then applied to a geographic database to create a predictive map. Kriging is a group of geostatistical techniques to spatially interpolate point values at an unobserved location from observations of values at nearby locations. The main problem with using kriging as an interpolator is that it is excessively data-dependent and requires a large number of closely spaced data points. Hence, there is a need to minimize the number of data points without sacrificing the accuracy of the results. In this paper, an Artificial Neural Networks (ANN) scheme was used to predict macronutrient values at un-sampled points. ANN has become a popular tool for prediction as it eliminates certain difficulties in soil property prediction, such as non-linear relationships and non-normality. Back-propagation multilayer feed-forward network structures were used to predict nitrogen, phosphorous and potassium values in the soil of the study area. A limited number of samples were used in the training, validation and testing phases of ANN (pattern reconstruction structures) to classify soil properties and the trained network was used for prediction. The soil analysis results of samples collected from the soil survey of block C of Sawah Sempadan, Tanjung Karang rice irrigation project at Selangor of Malaysia were used. Soil maps were produced by the Kriging method using 236 samples (or values) that were a combination of actual values (obtained from real samples) and virtual values (neural network predicted values). For each macronutrient element, three types of maps were generated with 118 actual and 118 virtual values, 59 actual and 177 virtual values, and 30 actual and 206 virtual values, respectively. To evaluate the performance of the proposed method, for each macronutrient element, a base map using 236 actual samples and test maps using 118, 59 and 30 actual samples respectively produced by the Kriging method. A set of parameters was defined to measure the similarity of the maps that were generated with the proposed method, termed the sample reduction method. The results show that the maps that were generated through the sample reduction method were more accurate than the corresponding base maps produced through a smaller number of real samples. For example, nitrogen maps that were produced from 118, 59 and 30 real samples have 78%, 62%, 41% similarity, respectively with the base map (236 samples) and the sample reduction method increased similarity to 87%, 77%, 71%, respectively. Hence, this method can reduce the number of real samples and substitute ANN predictive samples to achieve the specified level of accuracy.

Keywords: artificial neural network, kriging, macro nutrient, pattern recognition, precision farming, soil mapping

Procedia PDF Downloads 67

Authors: A. Delimitsou, C. Gouedard, E. Konstanta, A. Koletis, S. Patera, E. Manou, K. Spaho, S. Murray

Abstract:

Background: There remains a lack of International Standard Control Reference materials for Next Generation Sequencing-based approaches or device calibration. We have designed and validated dsDNA biomimetic reference materials for targeted such approaches incorporating proprietary motifs (patent pending) for device/test calibration. They enable internal single-sample calibration, alleviating sample comparisons to pooled historical population-based data assembly or statistical modelling approaches. We have validated such an approach for BRCA Copy Number Variation analytics using iQRS™-CNVSUITE versus Mixed Ligation-dependent Probe Amplification. Methods: Standard BRCA Copy Number Variation analysis was compared between mixed ligation-dependent probe amplification and next generation sequencing using a cohort of 198 breast/ovarian cancer patients. Next generation sequencing based copy number variation analysis of samples spiked with iQRS™ dsDNA biomimetics were analysed using proprietary CNVSUITE software. Mixed ligation-dependent probe amplification analyses were performed on an ABI-3130 Sequencer and analysed with Coffalyser software. Results: Concordance of BRCA – copy number variation events for mixed ligation-dependent probe amplification and CNVSUITE indicated an overall sensitivity of 99.88% and specificity of 100% for iQRS™-CNVSUITE. The negative predictive value of iQRS-CNVSUITE™ for BRCA was 100%, allowing for accurate exclusion of any event. The positive predictive value was 99.88%, with no discrepancy between mixed ligation-dependent probe amplification and iQRS™-CNVSUITE. For device calibration purposes, precision was 100%, spiking of patient DNA demonstrated linearity to 1% (±2.5%) and range from 100 copies. Traditional training was supplemented by predefining the calibrator to sample cut-off (lock-down) for amplicon gain or loss based upon a relative ratio threshold, following training of iQRS™-CNVSUITE using spiked iQRS™ calibrator and control mocks. BRCA copy number variation analysis using iQRS™-CNVSUITE™ was successfully validated and ISO15189 accredited and now enters CE-IVD performance evaluation. Conclusions: The inclusion of a reference control competitor (iQRS™ dsDNA mimetic) to next generation sequencing-based sequencing offers a more robust sample-independent approach for the assessment of copy number variation events compared to mixed ligation-dependent probe amplification. The approach simplifies data analyses, improves independent sample data analyses, and allows for direct comparison to an internal reference control for sample-specific quantification. Our iQRS™ biomimetic reference materials allow for single sample copy number variation analytics and further decentralisation of diagnostics to single patient sample assessment.

Keywords: validation, diagnostics, oncology, copy number variation, reference material, calibration

Procedia PDF Downloads 62

Authors: Zhidong Zhang, Zhi-Chao Zhang, Georgianna Duarte

Abstract:

This study investigated the attention problems with the instrument of Achenbach System of Empirically Based Assessment (ASEBA). Two thousand eight hundred and ninety-four adolescents were surveyed by using a stratified sampling method. We examined the relationships between relevant background variables and attention problems. Multiple regression models were applied to analyze the data. Relevant variables such as sports activities, hobbies, age, grade and the number of close friends were included in this study as predictive variables. The analysis results indicated that educational environments and extracurricular activities are important factors which influence students’ attention problems.

Keywords: adolescents, ASEBA, attention problems, educational environments, stratified sampling

Procedia PDF Downloads 274

Authors: Ella Tyuryumina, Alexey Neznanov

Abstract:

This paper is devoted to mathematical modelling of the progression and stages of breast cancer. We propose Consolidated mathematical growth model of primary tumor and secondary distant metastases growth in patients with lymph nodes metastases (CoM-III) as a new research tool. We are interested in: 1) modelling the whole natural history of primary tumor and secondary distant metastases growth in patients with lymph nodes metastases; 2) developing adequate and precise CoM-III which reflects relations between primary tumor and secondary distant metastases; 3) analyzing the CoM-III scope of application; 4) implementing the model as a software tool. Firstly, the CoM-III includes exponential tumor growth model as a system of determinate nonlinear and linear equations. Secondly, mathematical model corresponds to TNM classification. It allows to calculate different growth periods of primary tumor and secondary distant metastases growth in patients with lymph nodes metastases: 1) ‘non-visible period’ for primary tumor; 2) ‘non-visible period’ for secondary distant metastases growth in patients with lymph nodes metastases; 3) ‘visible period’ for secondary distant metastases growth in patients with lymph nodes metastases. The new predictive tool: 1) is a solid foundation to develop future studies of breast cancer models; 2) does not require any expensive diagnostic tests; 3) is the first predictor which makes forecast using only current patient data, the others are based on the additional statistical data. Thus, the CoM-III model and predictive software: a) detect different growth periods of primary tumor and secondary distant metastases growth in patients with lymph nodes metastases; b) make forecast of the period of the distant metastases appearance in patients with lymph nodes metastases; c) have higher average prediction accuracy than the other tools; d) can improve forecasts on survival of breast cancer and facilitate optimization of diagnostic tests. The following are calculated by CoM-III: the number of doublings for ‘non-visible’ and ‘visible’ growth period of secondary distant metastases; tumor volume doubling time (days) for ‘non-visible’ and ‘visible’ growth period of secondary distant metastases. The CoM-III enables, for the first time, to predict the whole natural history of primary tumor and secondary distant metastases growth on each stage (pT1, pT2, pT3, pT4) relying only on primary tumor sizes. Summarizing: a) CoM-III describes correctly primary tumor and secondary distant metastases growth of IA, IIA, IIB, IIIB (T1-4N1-3M0) stages in patients with lymph nodes metastases (N1-3); b) facilitates the understanding of the appearance period and inception of secondary distant metastases.

Keywords: breast cancer, exponential growth model, mathematical model, primary tumor, secondary metastases, survival

Procedia PDF Downloads 297

Authors: Gaurav Bisht, Rahul Rathnakumar, Ravikumar Duggirala

Abstract:

Predictive simulations are one of the key focus areas in safety-critical industries such as aerospace and high-performance automotive engineering. The stone-chipping study is one such effort taken up by the industry to predict and evaluate the damage caused due to gravel impact on vehicles. This paper describes a finite elements based method that can simulate the ejection of gravel chips from a vehicle tire. The FE simulations were used to obtain the initial ejection velocity of the stones for various driving conditions using a computational contact mechanics approach. To verify the accuracy of the tire model, several parametric studies were conducted. The FE simulations resulted in stone loft velocities ranging from 0–8 m/s, regardless of tire speed. The stress on the tire at the instant of initial contact with the stone increased linearly with vehicle speed. Mesh convergence studies indicated that a highly resolved tire mesh tends to result in better momentum transfer between the tire and the stone. A fine tire mesh also showed a linearly increasing relationship between the tire forward speed and stone lofting speed, which was not observed in coarser meshes. However, it also highlighted a potential challenge, in that the ejection velocity vector of the stone seemed to be sensitive to the mesh, owing to the FE-based contact mechanical formulation of the problem.

Keywords: abaqus, contact mechanics, foreign object debris, stone chipping

Procedia PDF Downloads 259

Authors: Zhiyuan Du, Baisravan Hom Chaudhuri, Pierluigi Pisu

Abstract:

In connected vehicle systems where wireless communication is available among the involved vehicles and intersection controllers, it is possible to design an intersection coordination strategy that leads the connected and automated vehicles (CAVs) travel through the road intersections without the conventional traffic light control. In this paper, we present a distributed coordination strategy for the CAVs at multiple interconnected intersections that aims at improving system fuel efficiency and system mobility. We present a distributed control solution where in the higher level, the intersection controllers calculate the road desired average velocity and optimally assign reference velocities of each vehicle. In the lower level, every vehicle is considered to use model predictive control (MPC) to track their reference velocity obtained from the higher level controller. The proposed method has been implemented on a simulation-based case with two-interconnected intersection network. Additionally, the effects of mixed vehicle types on the coordination strategy has been explored. Simulation results indicate the improvement on vehicle fuel efficiency and traffic mobility of the proposed method.

Keywords: connected vehicles, automated vehicles, intersection coordination systems, multiple interconnected intersections, model predictive control

Procedia PDF Downloads 353

Authors: Ibrahim Makram Ibrahim Salib

Abstract:

Agriculture plays an essential role in providing food for the world's population. It also offers numerous benefits to countries, including non-food products, transportation, and environmental balance. Precision agriculture, which employs advanced tools to monitor variability and manage inputs, can help achieve these benefits. The increasing demand for food security puts pressure on decision-makers to ensure sufficient food production worldwide. To support sustainable agriculture, unmanned aerial vehicles (UAVs) can be utilized to manage farms and increase yields. This paper aims to provide an understanding of UAV usage and its applications in agriculture. The objective is to review the various applications of UAVs in agriculture. Based on a comprehensive review of existing research, it was found that different sensors provide varying analyses for agriculture applications. Therefore, the purpose of the project must be determined before using UAV technology for better data quality and analysis. In conclusion, identifying a suitable sensor and UAV is crucial to gather accurate data and precise analysis when using UAVs in agriculture.

Keywords: agriculture land, agriculture land loss, Kabul city, urban land expansion, urbanization agriculture yield growth, agriculture yield prediction, explorative data analysis, predictive models, regression models drone, precision agriculture, farmer income

Procedia PDF Downloads 67