Abstract : Many predictive techniques have been widely applied in clinical decision Making such as predicting occurrence of a disease or diagnosis, evaluating Prognosis or outcome of diseases and assisting clinicians to recommend Treatment of diseases. However, the conventional predictive models or techniques are still not effective enough in capturing the underlying knowledge because it is incapable of simulating the complexity on feature representation of the medical problem domains. This research reports predictive analytical techniques for stroke diseases using deep learning model applied on heart disease dataset. The trtia fibrillation symptoms in heart patients are a major risk factor of stroke and share common variables to predict stroke. The outcomes of this research are more accurate than medical scoring systems currently in use for warning heart patients if they are likely to develop stroke.
 We apply the bi-CNN to estimate four perfusion parameters: CBV, CBF, T max, and MTT. Our results show that the bi-CNN estimations are comparable to the existing techniques, demonstrating that CNNs are capable of learning rich temporal feature filters that can extract important patterns from the data to make accurate parameter estimations. To the best of our knowledge, our work is the first to leverage deep learning techniques in the area ofperfusion parameter estimation
 ? The training set will compare between previous cases and new cases. However, diagnosis of disease is difficult problem because the number of risk factors are increase and complex. So, it need to improve prediction accuracy. ? In particular, there have been many computer-aided diagnosis systems using deep learning for detecting diverse diseases. Machine-learning/deep learning has been employed to detect or predict certain diseases using various approaches and datasets.
 ? There is limited previous work on utilizing machine learning algorithms to estimate perfusion parameters. In this work, we present a novel bi-input convolutional neural network (bi-CNN) to approximate four perfusion parameters without using an explicit deconvolution method. ? These bi-CNNs produced good approximations for all four parameters, with relative average root-mean-square errors (ARMSEs) =5% of the maximum values. ? We further demonstrate the utility of the estimated perfusion maps for quantifying the salvageable tissue volume in stroke, with more than 80% agreement with the ground truth. ? These results show that deep learning techniques area promising tool for perfusion parameter estimation without requiring a standard deconvolution process.
 ? The dataset we used included 13 variables, where the data were classified into training and testing data. ? We used both ABC and scaler to convert categorical variable in to continuous variable, and to generate models for testing. ? We then trained the DNN using ABC variables and compared to predict the results. ? The training and testing data did not overlap with one another. ? We demonstrate the predictive performance of our OCT models on the tasks of predictingin-hospital mortality, mortality within a year from discharge, and recurrence within a year fromdischarge.
Download DOC Download PPT

We have more than 145000 Documents , PPT and Research Papers

Have a question ?

Mail us : info@nibode.com