Lactated Ringers vs Normal Saline Resuscitation for Mild Acute Pancreatitis: A Randomized Trial (original) (raw)
Related papers
MCS-MCMC for Optimising Architectures and Weights of Higher Order Neural Networks
International Journal of Intelligent Systems and Applications
The standard method to train the Higher Order Neural Networks (HONN) is the well-known Backpropagation (BP) algorithm. Yet, the current BP algorithm has several limitations including easily stuck into local minima, particularly when dealing with highly non-linear problems and utilise computationally intensive training algorithms. The current BP algorithm is also relying heavily on the initial weight values and other parameters picked. Therefore, in an attempt to overcome the BP drawbacks, we investigate a method called Modified Cuckoo Search-Markov chain Monté Carlo for optimising the weights in HONN and boost the learning process. This method, which lies in the Swarm Intelligence area, is notably successful in optimisation task. We compared the performance with several HONN-based network models and standard Multilayer Perceptron on four (4) time series datasets: Temperature, Ozone, Gold Close Price and Bitcoin Closing Price from various repositories. Simulation results indicate tha...
International Conference on Emerging Applications and Technologies for Industry 4.0 (EATI’2020), 2021
For the data classification task back propagation (BP) is the most common used model to trained artificial neural network (ANN). Various parameters were used to enhance the learning process of this network. However, the conventional algorithms have some weakness, during training. The error function of this algorithm is not explicit to locate the global minimum, while gradient descent may cause slow learning rate and get stuck in local minima. As a solution, nature inspired cuckoo search algorithms provide derived free solution to optimize composite problems. This paper proposed a novel meta-heuristic search algorithm, called cuckoo search (CS), with variable learning rate to train the network. The proposed variable learning rate with cuckoo search algorithm speed up the slow convergence and solve the local minima problem of the backpropagation algorithm. The proposed CS variable learning rate BP algorithms are compared with traditional algorithms. Particularly, diabetes and cancer benchmark classification problems datasets are used. From the analyses results it show that proposed algorithm shows high efficiency and enhanced performance of the BP algorithm.
Optimization of Neural Network using Cuckoo Search for the Classification of Diabetes
Available records show that over 80% of the patient suffering from diabetes die from heart or blood diseases. Total cure for the diabetes is currently not available. In this paper, we proposed diabetes classifier based on the cuckoo search algorithm (CS) and Neural Network (NN). The weights and bias of the NN was trained using the CS to deviate from being stuck in local minima. The high dimension of the features in our dataset triggered the study to extract the critical features using principal component analysis. The extracted features were used to built a classifier based on the NN and the CS for classifying potential diabetes patients. The propose diabetes classifier performance was compared to the classifiers built based on artificial bee colony and genetic algorithm. Simulation results show that the proposed approach converges faster to the optimum solution than the comparative classifiers. Comparative study of the approach proposed and previous methods, further proved the effectiveness of our method. The classifier has provided promising classification result in the classifying of potential diabetic patients. The classifier have the capability of automatically diagnosing possible diabetic patients. This can be of help to the physicians in taken decision about the status of a diabetic patient.
This paper aims to investigate the ability of the proposed Modified Cuckoo Search-Markov chain Monte Carlo (MCS-MCMC) algorithm for two (2) types of Higher Order Neural Networks (HONNs); Pi-Sigma Neural Networks and Functional Link Neural Networks that will influence the performance of searching ability, even at high numbers of dimensions. We validated the proposed MCS-MCMC algorithm alongside several benchmark test functions The proposed MCS-MCMC were tested on three (3) different time-series data; relative humidity, temperature and laser datasets. The performance of those HONNs is benchmarked against the performance of Multilayer Perceptron. The simulation results shows that, by incorporating MCS-MCMC algorithm in both HONNs can improve convergence rate and decrease the prediction error.
A New Back-Propagation Neural Network Optimized with Cuckoo Search Algorithm
Springer-Verlag Berlin Heidelberg, 2013
Back-propagation Neural Network (BPNN) algorithm is one of the most widely used and a popular technique to optimize the feed forward neural network training. Traditional BP algorithm has some drawbacks, such as getting stuck easily in local minima and slow speed of convergence. Nature inspired meta-heuristic algorithms provide derivative-free solution to optimize complex problems. This paper proposed a new meta-heuristic search algorithm, called cuckoo search (CS), based on cuckoo bird’s behavior to train BP in achieving fast convergence rate and to avoid local minima problem. The performance of the proposed Cuckoo Search Back-Propagation (CSBP) is compared with artificial bee colony using BP algorithm, and other hybrid variants. Specifically OR and XOR datasets are used. The simulation results show that the computational efficiency of BP training process is highly enhanced when coupled with the proposed hybrid method.
A Modified Weight Optimization for Artificial Higher Order Neural Networks in Physical Time Series
International Journal of Advanced Computer Science and Applications
Many methods and approaches have been proposed for analyzing and forecasting time series data. There are different Neural Network (NN) variations for specific tasks (e.g., Deep Learning, Recurrent Neural Networks, etc.). Time series forecasting are a crucial component of many important applications, from stock markets to energy load forecasts. Recently, Swarm Intelligence (SI) techniques including Cuckoo Search (CS) have been established as one of the most practical approaches in optimizing parameters for time series forecasting. Several modifications to the CS have been made, including Modified Cuckoo Search (MCS) that adjusts the parameters of the current CS, to improve algorithmic convergence rates. Therefore, motivated by the advantages of these MCSs, we use the enhanced MCS known as the Modified Cuckoo Search-Markov Chain Monté Carlo (MCS-MCMC) learning algorithm for weight optimization in Higher Order Neural Networks (HONN) models. The Lévy flight function in the MCS is replaced with Markov Chain Monté Carlo (MCMC) since it can reduce the complexity in generating the objective function. In order to prove that the MCS-MCMC is suitable for forecasting, its
Improved Cuckoo Search Algorithm for Feed forward Neural Network Training
International Journal of Artificial Intelligence & Applications, 2011
The cuckoo search algorithm is a recently developed meta-heuristic optimization algorithm, which is suitable for solving optimization problems. To enhance the accuracy and convergence rate of this algorithm, an improved cuckoo search algorithm is proposed in this paper. Normally, the parameters of the cuckoo search are kept constant. This may lead to decreasing the efficiency of the algorithm. To cope with this issue, a proper strategy for tuning the cuckoo search parameters is presented. Then, it is employed for training feedforward neural networks for two benchmark classification problems. Finally, the performance of the proposed algorithm is compared with that of the standard cuckoo search. Simulation results demonstrate the effectiveness of the proposed algorithm.
This paper demonstrates that neural nets have the capacity to 'mould' themselves to data sets which relate to critically ill patients. Furthermore, they outperformed the conventional approach of logistic regression analysis and were successfully 'bred' for improved performance using genetic algorithms. Neural net analysis of databases relating to ICUs has already been performed by others. Doig et al [29] used 15 physiological parameters measured on 422 ICU patients to train a neural net for mortality-rate prediction. Whilst their performance figures were an apparent improvement over those obtained from a logistic regression model, the low frequency of mortality in their dataset precluded a valid comparison between the two methods. Buchman et al [30] also compared the performance of a neural net with a logistic regression model with regard to predicting the length of stay in ICU. The network, constructed using daily records from from 491 patients, outperformed the con...
Mortality Prediction in ICU Using Artificial Neural Networks
International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2022
Intensive Care Unit (ICU) death rate prediction supported artificial neural network, patients WHO have important unwellness and injury can get admitted to the unit. The dying rate for sufferers admitted to the unit can vary from Associate in Nursing underlying illness with a dying rate as low as one in twenty sufferers admitted for non-compulsory surgical operation and as excessive as one in 4 sufferers with metabolic process diseases. Artificial Neural Networks facilitate health care management selections, improve care and cut back value at same time by exploiting the applications of ANN within the care department. It helps to alter, minimize errors and predict additional correct diagnoses for the diseases or injuries that may cut back patients' risk of hospitalization supported given knowledge. Associate in Nursing empiric study was conducted with 4000+ adult patients from 100+ ICUs. The patients with ≥1 organ failure are forty sixth. Patients without infection receiving antibiotics stand for an hour. There have been 500+ deaths and 183 discharges from the unit. In 1627 patients admitted among twenty four h of the study day, the standardized mortality quantitative relation was zero.67. We have a tendency to develop our model exploitation of a synthetic Neural Network to predict the death rate of sick patients supported by their diseases, injuries, and medical records. This can additionally facilitate managing the unit beds in hospitals throughout the Associate in Nursing emergency.
Prediction Using Cuckoo Search Optimized Echo State Network
Arabian Journal for Science and Engineering, 2019
The advent of internet of things has brought a revolution in the amount of data generated in industry. Researchers now have to develop ways to harness such huge amount of data. Thus, a new method called "predictive maintenance" was developed. In this technique, sensor data is used to predict failures so that appropriate actions are taken to save accidents and costs. Artificial neural networks have proven to be excellent tools for prediction. In this work, the echo state network (ESN), which is a new concept of recurrent neural network (RNN), is used to predict failures in turbofan engines. The ESN was developed to solve the complexities of earlier RNNs. However, choosing the right topology and parameters for the ESN is often a difficult problem. Hence, we develop a cuckoo search optimization-based algorithm to optimize the ESN. The approach is compared with three particle swarm optimization methods and two other methods, and it performed better.