Proposed Prediction System Research Articles

Predicting the progression of mild cognitive impairment to Alzheimer’s disease by longitudinal magnetic resonance imaging-based dictionary learning

ObjectiveEfficient prediction of the progression of mild cognitive impairment (MCI) to Alzheimer’s disease (AD) is important for the early intervention and management of AD. The aim of our study was to develop a longitudinal structural magnetic resonance imaging-based prediction system for MCI progression. MethodsA total of 164 MCI patients with longitudinal data were collected from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). After preprocessing, a discriminative dictionary learning framework was applied to differentiate MCI patches, avoiding the segmentation of regions of interest. Then, the proportion of patches classified as more severe atrophy patches in a patient was calculated as his or her feature to be input into a simple support vector machine. Finally, a new subject was predicted with fourfold cross-validation (CV), and the area under the receiver operating characteristic curve (AUC) was determined. ResultsThe average accuracy and AUC values after fourfold CV were 0.973 and 0.984, respectively. The effects of the data from one or two time points were also investigated. ConclusionThe proposed prediction system achieves desirable and reliable performance in predicting progression for MCI patients. Additionally, the prediction of MCI progression with longitudinal data was more effective and accurate. SignificanceThe developed scheme is expected to advance the clinical research and treatment of MCI patients.

An effective disease prediction system using incremental feature selection and temporal convolutional neural network

Rapid growth of communication technologies and expert systems produces enormous volume of medical data. Deep learning technique is an advancement of machine learning technique for analysing the huge amount of various diseases related medical dataset. Even though, no healthcare systems are achieved better prediction accuracy with the various medical datasets in the past decades. For improving the accuracy level of disease prediction, we develop a disease prediction system to predict the serious death diseases including heart disease, diabetic disease and cancer diseases effectively in this paper. This disease prediction system consists of feature selection method that works as incremental in nature named as Incremental Feature Selection Algorithm (IFSA) which combines the concepts of Intelligent Conditional Random Field (CRF) on feature selection process and the Linear Correlation Coefficient based Feature Selection (ICRF-LCFS) method algorithm and an existing Convolutional Neural Network (CNN) with temporal features (T-CNN). The proposed disease prediction system is evaluated and achieved better prediction accuracy in less time with low false alarm rate.

PREDICTION MODEL OF WIND SPEED AND DIRECTION USING DEEP NEURAL NETWORK

This paper presents the prediction system of wind speed and direction using a feed-forward backpropagation neural network (FFBPNN). The input of the prediction system is wind speed and direction which are numerical data and provided by Automated Meteorological Data Acquisition System (AMeDAS) in Japan. The performances of the proposed system is evaluated based on mean square error (MSE) between predicted and observed data. In this paper, we substantiate the usefulness of the proposed prediction system improving prediction accuracy compared to four prediction models.

Assessment of Classifiers with Machine Learning based Thyroid Disease Prediction System

Most of the people in different nations are suffering from Thyroid related diseases and these are lifelong. Many people are unaware of having Thyroid related diseases. Main cause for this is due to improper functioning of Thyroid gland secreting Thyroid hormone which regulates body metabolism. In this paper we have made survey on classifiers like Decision Tree C4.5(J48), Multilayer Perceptron, Naïve Bayes by measuring TP Rate, FP Rate, Precision, Recall, F-Measure, MCC, ROC Area, PRC Area and developed a prediction system for Thyroid diseases. For training and testing the classifiers we have used Thyroid dataset from UCI repository. Dataset consists of 9172 records containing 29 attribute values and 1 diagnosis class value. The diagnosis class value consists of different types Thyroid disease conditions like hyperthyroid conditions, hypothyroid conditions, binding protein, general health, replacement therapy, antithyroid treatment and miscellaneous. The proposed prediction system model capable of predicting type of Thyroid disease whether a person is suffering or not.

Discrimination of Golgi Proteins Through Efficient Exploitation of Hybrid Feature Spaces Coupled With SMOTE and Ensemble of Support Vector Machine

Many organelles inside and outside a living cell depend on the perfect behavior of Golgi apparatus for smooth and normal functioning. Its poor performance may lead to many inheritable diseases like diabetes and cancer. Therefore, it is highly crucial to detect any strange behavior of Golgi apparatus in advance. Accurate discrimination of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cis</i> -Golgi from <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">trans</i> -Golgi proteins surely helps researchers identify the role of Golgi proteins in various diseases and assist pharmacists in drug development. In this work, various hybrid models of Bi-Profile Bayes, Bigram PSSM, Di-Peptide Composition, and Split Amphiphilic Pseudo Amino Acid Composition with SMOTE oversampling technique have been employed to discriminate Golgi protein types. Multiple linear Support Vector Machines have been used to exploit the discrimination power of these models. The proposed prediction system: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Golgi-predictor</i> has shown significant performance and achieved promising results compared to other existing state-of-the-art techniques. Through the 10-fold cross-validation, the proposed system achieved an accuracy value of 97.6%, sensitivity value of 98.8%, specificity value of 96.5%, G-mean value of 97.6%, MCC value of 0.95, and F-score value of 0.97. Similarly, through the jackknife cross-validation, the achieved values for accuracy, sensitivity, specificity, G-mean, MCC, and F-score are respectively, 96.5%, 97.8%, 95.2%, 96.4%, 0.93, and 0.96. Moreover, through the independent dataset testing, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Golgi-predictor</i> demonstrated significant enhancement in performance over other techniques. The proposed methodology aims at supporting drug designers in pharmaceutical industry and assisting researchers from the fields of bioinformatics and computational biology towards better innovation in predicting the behavior of Golgi proteins.

Big Data Based Diabetes and Heart Disease Prediction System by Employing Supervised Learning Algorithm

Healthcare systems generate bytes and bytes of data and the data growth is exponential. The voluminous data can be analysed effectively, only when the data organization is efficient. Additionally, data retrieval must also be made simpler, such that the healthcare professional can compare and contrast the test sample with the database of health records. This makes it possible to achieve better disease prediction and this work presents a big data based disease prediction system with the help of supervised learning. The proposed approach clusters the related health records, based on every medical attribute followed by which the disease is predicted by SVM classifier. The performance of the proposed disease prediction system is observed to be satisfactory in terms of accuracy, precision, recall, F-measure, while consuming reasonable period of time.

Hybrid user action prediction system for automated home using association rules and ontology

Nowadays, with the rapid increase of Internet users, the Internet services dominate a primary part of our lifestyle. Moreover, the evolution of the internet of things has introduced new insights into smart platforms and devices that leads to the new vision of ‘smart homes’. The idea of smart homes is not a recent concept; it has been in high interest by both academia and industry to make smart homes a more convenient technology for human's comfort. In this study, the authors propose a new hybrid prediction system based on the frequent pattern (FP)-growth and ontology graphs for home automation systems. Their proposed system simulates the human prediction actions by adding common sense data by utilizing the advantages of the ontology graph and the FP-growth to find a better solution in predicting home user actions for automated systems. For the evaluation of the proposed system, two ontology graphs are introduced with FP-growth to achieve the best results. Both graphs are tested through multiple weight values with the results of FP-growth. As a result, the best weight distribution selected in this study is (70, 30) for time and location ontology graphs respectively. Their results showed that the proposed prediction system achieved an accuracy of 79% for all weekdays and 81% excluding weekend days.

Regression-based software fault prediction using biogeography-based optimisation (R-BBO)

It is difficult to build model of accurate estimate due to the inherent uncertainty and similarity among different categories in development projects. In this paper, fault prediction is done using biogeography-based optimisation (BBO) with the goal of recognising the faults in software systems in more efficient way. Our methodology includes four steps as follows: 1) firstly pre-processing was employed to remove redundant data; 2) secondly, relevant features are extracted using principal component analysis; 3) thirdly, fault-prediction system based on the optimisation of regression parameter using biogeography-based optimisation (R-BBO) was proposed. The experiment employed over different fault related datasets using ten-fold cross validation. The results showed that proposed prediction system (R-BBO) yield an overall accuracy of 85.4% (predicted over five datasets) which is higher than the prediction using genetic algorithm (R-GA). The proposed R-BBO was effective in terms of classification accuracy, precision and recall.

Regression-based software fault prediction using biogeography-based optimisation (R-BBO)

It is difficult to build model of accurate estimate due to the inherent uncertainty and similarity among different categories in development projects. In this paper, fault prediction is done using biogeography-based optimisation (BBO) with the goal of recognising the faults in software systems in more efficient way. Our methodology includes four steps as follows: 1) firstly pre-processing was employed to remove redundant data; 2) secondly, relevant features are extracted using principal component analysis; 3) thirdly, fault-prediction system based on the optimisation of regression parameter using biogeography-based optimisation (R-BBO) was proposed. The experiment employed over different fault related datasets using ten-fold cross validation. The results showed that proposed prediction system (R-BBO) yield an overall accuracy of 85.4% (predicted over five datasets) which is higher than the prediction using genetic algorithm (R-GA). The proposed R-BBO was effective in terms of classification accuracy, precision and recall.

Wind Speed Forecasting System Based on the Variational Mode Decomposition Strategy and Immune Selection Multi-Objective Dragonfly Optimization Algorithm

In the development of the wind power industry, short-term wind speed forecasting is necessary, and many researchers have made substantial efforts to establish wind speed prediction models. However, realizing the accurate prediction of wind speeds remains a challenging task. The current prediction models do not consider the preprocessing of the data, and each model has various shortcomings. Considering the disadvantages of the available models, in this paper, an advanced combined forecasting system is applied that utilizes a data preprocessing strategy and parameter optimization strategy to obtain accurate prediction values. The proposed prediction system employs linear and nonlinear models that can take into account the characteristics of wind speed sequences, successfully combine the advantages of various single models, and yield accurate and stable prediction values. Finally, according to the experimental analysis and discussion, the proposed combined prediction system outperforms the compared models in prediction. In conclusion, the powerful combined prediction model provides a feasible scheme for wind power prediction.

Load level prediction system model of friction stir spot welded aluminium alloy using support vector machine

The cost and efficiency of experiment and test are still a major issue in manufacturing especially in welding. In this study, a machine learning technique i.e. support vector machine (SVM) was applied to develop a load level prediction system of friction stir spot welded joint aluminium alloy AA5052-H112. This load level prediction system model was proposed based on three levels of load group of the welded joint. Experimental works in friction stir spot welding was conducted on samples specimens of AA5052-H112 2mm thick overlap joint based on 27 combinations of governed parameters i.e. spindle speed, tool depth, and dwell time. Mechanical testing of tensile shear load was then carried out to those specimens to obtain 27 loads data to form the SVM classifier. These data were required for pattern classification and model development via training and testing the proposed prediction system. The result obtained via training and testing showed the classification of load data produced by the proposed system, matched to the required load level with the 100%. This study proved that the proposed load level prediction system offered a useful tool to predict the load level of friction stir welded joint aluminium alloy AA5052-H112 by using respected parameters without required experiments and tests.

NON-EUCLIDEAN, CONVOLUTIONAL LEARNING ON CORTICAL BRAIN SURFACES.

In recent years there have been many studies indicating that multiple cortical features, extracted at each surface vertex, are promising in the detection of various neurodevelopmental and neurodegenerative diseases. However, with limited datasets, it is challenging to train stable classifiers with such high-dimensional surface data. This necessitates a feature reduction that is commonly accomplished via regional volumetric morphometry from standard brain atlases. However, current regional summaries are not specific to the given age or pathology that is studied, which runs the risk of losing relevant information that can be critical in the classification process. To solve this issue, this paper proposes a novel data-driven approach by extending convolutional neural networks (CNN) for use on non-Euclidean manifolds such as cortical surfaces. The proposed network learns the most powerful features and brain regions from the extracted large dimensional feature space; thus creating a new feature space in which the dimensionality is reduced and feature distributions are better separated. We demonstrate the usability of the proposed surface-CNN framework in an example study classifying Alzheimers disease patients versus normal controls. The high performance in the cross-validation diagnostic results shows the potential of our proposed prediction system.

Development of the system for prediction of security state of an enterprise using semantic–frame fuzzy models of knowledge base

The system based on results of prediction of technical and technological potential was proposed; it provided clear current values of indicators of the enterprise’s operation, properties and permissible restrictions of these properties. It was proved that the proposed system has a special purpose and consists of units of classes of indicators and the base of the facts of individual instances of classes, and can be used for the implementation of prediction problems in any industry. As a result of modeling, it was found that introduction of the prediction unit into a system allows us more accurately and objectively to consider and evaluate a whole range of indicators of the enterprise’s operation. The proposed prediction system calculated approximated prospective value of the indicator of the state of technical and technological potential of an enterprise in time, which greatly affects probability of bankruptcy of an enterprise. It is appropriate to use the prediction system for complex processes with fuzzy logic, when there is no simple mathematical model and expert knowledge can be formulated without fuzzy logic only in linguistic form. This proves that the proposed system can be used for prediction of all other potential of an enterprise that also influence probability of bankruptcy of an enterprise.

PREDICTION OF UNFAVORABLE COURSE OF GLOMERULONEPHRITIS WITH NEPHROTIC SYNDROME IN CHILDREN: ENZYMURIC CRITERIA

The aim of study was to investigate the possibility of using the detection of urine renospecific enzymes activity in children with nephrotic syndrome (NS) as biomarkers of unfavorable disease course prognosis. Materials and methods. The operating characteristics of urine activity values of N-аcethyl- β-D-glucosaminidase (NAG) and β-galactosidase (β-Gal) were estimated in 209 nephrotic children depending on disease course. Results. The high activity of NAG and β-Gal confirms the active participation of renal tubules in pathological process in NS. Likelihood of NS progression increased while saving NAG activity &gt;50 Units and β-Gal &gt;30 Units in 8-10 weeks of therapy and at 1.5-3 years of follow-up after treatment withdrawal; increasing of enzymes levels in 6-9 months on tapering treatment period (NAG &gt;100 Units, β-Gal &gt;60 Units). Changing the enzymes activity in the disease active stage before treatment does not have the prognostic defense. The enzymuric fragment of multimarker parallel cascade system for prediction unfavorable NS course has been created. Conclusion. Urine renospecific enzymes activity (NAG and β-Gal) should be used as markers of the progression in NS children with the optimal terms studies and range parameters as summarized in the proposed prediction system.

The first step towards a respiratory motion prediction for natural-breathing by using a motion generator

Respiratory gated radiation therapy (RGRT) gives accurate results when a patient’s breathing is stable and regular. Thus, the patient should be acutely aware during respiratory pattern training before undergoing the RGRT treatment. To bypass the process of respiratory pattern training, we propose a tumor location prediction system for RGRT that uses only the natural respiratory volume, and we confirm its application. In order to verify the proposed tumor location prediction system, we used an in-house phantom set. The set involved a chest phantom with target, external markers and a motion generator. Natural respiratory volume signals were generated using the random function in the MATLAB code. In the chest phantom, the target undergoes linear motion based on the respiratory signal. After a four-dimensional computed tomography (4DCT) scan of the in-house phantom, the motion trajectory was derived as a linear equation. The accuracy of the linear equation was compared with that of the motion algorithm used by the operating motion generator. In addition, we attempted to predict the tumor’s location by using the random respiratory volume values. The correspondence rate of the linear equation derived from the 4DCT images with the motion algorithm of the motion generator was 99.41% (p > 0.05). Also, the average error rate of the tumor-location prediction was 1.23% for 26 cases. We confirmed the applicability of our proposed tumor location prediction system using the natural respiratory volume for RGRT. If additional clinical studies can be conducted, a more accurate prediction that would not require respiratory pattern training can be realized.

Automated Forecasting Approach Minimizing Prediction Errors of CPU Availability in Distributed Computing Systems

Forecasting CPU availab ility in volunteer computing systems using a single prediction algorith m is insufficient due to the d iversity of the world -wide distributed resources. In this paper, we draw-up the main guidelines to develop an appropriate CPU availability prediction system fo r such computing infrastructures. To reduce solution time and to enhance precision, we use simp le pred iction techniques, precisely vector autoregressive models and a tendency-based technique. We propose a predictor construction process which automatically checks assumptions of vector autoregressive models in time series. Three different past analyses are performed. For a given volunteer resource, the proposed prediction system selects the appropriate predictor using the mult i-state based prediction technique. Then, it uses the selected predictor to forecast CPU availability indicators. We evaluated our predict ion system using real traces of more than 226000 hosts of Seti@ho me. We found that the proposed prediction system improves the prediction accuracy by around 24%. Index Terms—CPU availab ility prediction, predict ion system, mult ivariate time series, mult i-state based prediction, volunteer computing system.

Resource Allocation With Video Traffic Prediction in Cloud-Based Space Systems

This paper considers the resource allocation problems for video transmission in space-based information networks. The queueing system analyzed in this study is constituted by multiple users and a single server. The server is operated as a cloud that can sense the traffic arrivals to each user's queue and then allocates the transmission resource and service rate for users. The objectives are to make configurations over time to minimize the time average cost of the system, and to minimize the waiting time of packets after they enter the queue. Meanwhile, the constraints on the queue stability of the system must be satisfied. In this paper, we introduce a predictive backpressure algorithm, which considers the future arrivals with a certain prediction window size into the consideration of resource allocation to make decisions on which packets to be served first. In addition, this paper designs a multiresolution wavelet decomposition-based backpropagation network for the prediction of video traffic, which exhibits the long-range dependence property. Simulation results indicate that the delay of the queueing system can be reduced through this prediction-based resource allocation, and the prediction accuracy for the video traffic is improved according to the proposed prediction system.

Protein subcellular localization of fluorescence microscopy images: Employing new statistical and Texton based image features and SVM based ensemble classification

Image based subcellular localization of proteins is a challenging pattern recognition task and is of prime importance in understanding different functions of proteins. Extracting information about the protein's subcellular location using discriminative feature extraction strategies helps in understanding the behavior of different proteins under different circumstances.In this work, an intelligent prediction system for protein subcellular localization using fluorescence microscopy images is developed. The proposed prediction system IEH-GT uses a new feature extraction strategy and ensemble classification. The new feature extraction mechanism exploits statistical and Texton based image descriptors, whereas ensemble classification is performed using the majority voting based ensemble of SVMs. The novelty of IEH-GT prediction system lies (a) in the individual exploitation of the feature spaces generated from both individual Gray Level Co-occurrence Matrices (GLCMs) and Texton images as well as (b) in the manner the extracted features are exploited using the learning capabilities of SVM, which is utilized as base classifier in majority voting based ensemble. In this work, the effect of the individual orientations of GLCMs and Texton images on the capability of protein-recognition is investigated.It is observed that GLCMs along different orientations characterize different information regarding intensity distribution and relative positions of pixels in a certain neighborhood. Similarly, the individually generated features from each Texton image help in reducing the overlapping information. The proposed IEH-GT prediction system is thus able to effectively exploit information about the patterns of intensity-variations of proteins. It is experimentally observed that due to the capability of the proposed IEH-GT system in differentiating diverse surface structures of proteins, the mis-predictions of similar proteins such as Golgpp and Golgia are substantially reduced.

Wind power prediction system for wind farm based on auto regressive statistical model and physical model

Extracting energy from renewable sources such as wind energy is widely investigated in the past decades to mitigate the global energy crisis and environmental pollution. For a wind farm that converts wind energy into electricity power, a real-time prediction system of the output power is significant. In this paper, a prediction system is developed with a method of combining statistical model and physical model. In this system, the inlet condition of the wind farm is forecasted by the auto regressive model. The flow field is computed by the Reynolds average Navier-stokes simulation in the computational fluid dynamics model. The wake flow is calculated by the particle model, which can be used over complex terrain. Taking also the terrain condition, the property of turbines and wake flow model into account, the output power of the wind farm can be further predicted. The proposed prediction system is tested by the data from Wattle Point Wind Farm in Australia. Through the data post-processing, the error of the mean daily output power is less than 5%. The proposed system is effective for power output prediction of wind farm.

Prediction of protein structure classes using hybrid space of multi-profile Bayes and bi-gram probability feature spaces

Proteins are the executants of biological functions in living organisms. Comprehension of protein structure is a challenging problem in the era of proteomics, computational biology, and bioinformatics because of its pivotal role in protein folding patterns. Owing to the large exploration of protein sequences in protein databanks and intricacy of protein structures, experimental and theoretical methods are insufficient for prediction of protein structure classes. Therefore, it is highly desirable to develop an accurate, reliable, and high throughput computational model to predict protein structure classes correctly from polygenetic sequences. In this regard, we propose a promising model employing hybrid descriptor space in conjunction with optimized evidence-theoretic K-nearest neighbor algorithm. Hybrid space is the composition of two descriptor spaces including Multi-profile Bayes and bi-gram probability. In order to enhance the generalization power of the classifier, we have selected high discriminative descriptors from the hybrid space using particle swarm optimization, a well-known evolutionary feature selection technique. Performance evaluation of the proposed model is performed using the jackknife test on three low similarity benchmark datasets including 25PDB, 1189, and 640. The success rates of the proposed model are 87.0%, 86.6%, and 88.4%, respectively on the three benchmark datasets. The comparative analysis exhibits that our proposed model has yielded promising results compared to the existing methods in the literature. In addition, our proposed prediction system might be helpful in future research particularly in cases where the major focus of research is on low similarity datasets.