A Probabilistic Neural Network for Assessment of the Vesicoureteral Reflux’s Diagnostic Factors Validity

Vesicoureteral reflux detection with reliable probabilistic outputs

“…The only study we were able to find other than the ones of the Democritus University of Thrace group on the prediction of VUR using Machine Learning techniques is [3]. In this study the authors identified a urinary proteome pattern for detecting High-grade VUR (grade IV or V) with the use of sup- The studies [8,9,10] were performed on the same dataset we use here, however only a particular part of it consisting of 20% of the cases was used for testing. In [10] the authors studied the use of multilayer perceptrons, in [9] the authors studied the use of probabilistic neural networks (PNNs) and in [8] the PNNs were combined with a genetic algorithm for optimization of the feature subset and parameters used.…”

“…In this study the authors identified a urinary proteome pattern for detecting High-grade VUR (grade IV or V) with the use of sup- The studies [8,9,10] were performed on the same dataset we use here, however only a particular part of it consisting of 20% of the cases was used for testing. In [10] the authors studied the use of multilayer perceptrons, in [9] the authors studied the use of probabilistic neural networks (PNNs) and in [8] the PNNs were combined with a genetic algorithm for optimization of the feature subset and parameters used. The best results were achieved in [8], with a best accuracy of 96.3% on the particular test set.…”

“…The use of machine learning techniques towards this goal using as inputs clinical and laboratorial information that can be obtained without the need of radiation exposure was examined in some recent studies [3,8,9,10]. The techniques proposed in these studies however only provide a yes or no output, without giving any further information about how much one can rely on this output.…”

Artificial Neural Networks for Estimation of Dementias Types

“…ANNs are suitable for the prognosisof the diseaseas they use a set of examples that represent variants of the disease, instant of diagnostic rules. The neural networks have proven to be a powerful tool in the diagnosis of diseasesin different medical areas [4,6,[14][15][16][17], such as oncology [18][19][20][21], endocrinology [22][23][24], urology [25,26], pediatric [27], cardiology [28,29], radiology [30], ophthalmology [31], neurology [32][33][34] and others [35,36].…”

Solar radiation: Cloudiness forecasting using a soft computing approach

“…The ANNs are used in ample scope applications, such as weather prediction [12][13][14] , heat transfer prediction [15] , short term load forecasting [16][17][18][19] , numerical simulation of nonlinear equations [20,21] , financial prediction [22,23] , telecommunications, medicine [24,25] , signal processing and other fields [26][27][28] .…”