Real time decision support system for diagnosis of rare cancers, trained in parallel, on a graphics processing unit

Intelligent Decision Support Systems

“…Second, neural networks also can be viewed as a class of quantitative models to be used in a model-driven DSS (Delen and Sharda 2008). DSS employing a neural network are being developed in various areas of human activity (Sidiropoulos et al 2012;Tsadiras et al 2013;Arsene et al 2012;Azadeh et al 2012;Delen and Sharda 2008;Mendyk et al 2013;Abpeykar and Ghatee 2014;Srinivasan et al 2010). Next there is a short description of several DSS that were first mentioned as employing a neural network.…”

“…A new strategy is introduced by Sidiropoulos et al (2012) for designing and developing of an efficient dynamic Decision Support System (DSS) for supporting rare cancers decision making. The proposed DSS operates on a Graphics Processing Unit (GPU) and it is capable of adjusting its design in real time based on user-defined clinical questions in contrast to standard CPU implementations that are limited by processing and memory constrains.…”

A clinical decision support system for diagnosis of Allergic Rhinitis based on intradermal skin tests

“…The development in computing and technology, has led to the use of Clinical Decision Support System (CDSS) in medical centres and hospitals for efficient and reliable diagnosis and prognosis. Furthermore CDSS are used for Dental education [12], Ventilator treatment [13], Cancer Diagnosis [14], Cancer Treatment [15], Diagnosis of heart diseases [16] and Heart Failure [17], Diagnosis of neuromuscular disorders [18] Respiratory pattern analysis [19] and many more. The CDSS in this work uses a rule based knowledge mining approach, for detection of Allergic Rhinitis from the results of the Intradermal Skin Tests.…”

A pattern recognition system for prostate mass spectra discrimination based on the CUDA parallel programming model

“…To deal with the high-dimensionality problem of the PR-system design, most of the system design was transferred onto the processors of a GPU (GeForce GTX 580), attached to the computer, by adopting parallel processing programming using the CUDA toolkit v4.0 and the C/C++ programming language [9]. Figure 1 shows the parallelization of the training process, where in each GPU thread an mz-interval combination was fed, the PNN classifier was designed, trained and was evaluated for overall accuracy by means of LOO method.…”

“…In the present study, we propose a pattern- recognition (PR) system to discriminate between healthy and malignant prostate MS-spectra. The implementation and evaluation of the PR-system were performed by means of a graphics processing unit (GPU) framework using the parallel architecture of the GPU card and the Compute Unified Device Architecture (CUDA) programming environment [9] in order to attain optimal classifier design, minimizing the loss of candidate biomarkers.…”