The classification and recognition of two-dimensional patterns independently of their position, orientation, and size by using high-order networks are discussed. A method is introduced for reducing and controlling the number of weights of a third-order network used for invariant pattern recognition. The method leads to economical networks that exhibit high recognition rates for translated, rotated, and scaled, as well as locally distorted, patterns. The performance of these networks at recognizing types and handwritten numerals independently of their position, size, and orientation is compared with and found superior to the performance of a layered feedforward network to which image features extracted by the method of moments are presented as input.
Magnetic resonance spectroscopy (MRS) provides a non-invasive measurement of the biochemistry of living tissue. However, signal variation due to tissue heterogeneity causes considerable mixing between different disease categories, making accurate class assignments difficult. This paper compares a systematic methodology for classifier design using multivariate bayesian variable selection (MBVS), with one based on feature extraction using independent component analysis (ICA). We illustrate the methodology and assess the classification performance using a data set comprising 41 magnetic resonance spectra acquired in vivo from two grades of brain tumour, namely low- and medium-grade astrocytic tumours, labelled astrocytomas (AST), and high-grade gliomas and glioblastomas labelled glioblastomas (GL). The aim of this study is threefold. First, to describe the application of the alternative methodologies to MRS, then to benchmark their classification performance, and finally to interpret the classification models in terms of biologically relevant signals derived from the spectra. The classification performance is assessed using the bootstrap method and by application to a test sample in a retrospective study.
The aim of this study was to compare multilayer perceptron neural networks (NNs) with standard logistic regression (LR) to identify key covariates impacting on mortality from cancer causes, disease-free survival (DFS), and disease recurrence using Area Under Receiver-Operating Characteristics (AUROC) in breast cancer patients. From 1996 to 2004, 2,535 patients diagnosed with primary breast cancer entered into the study at a single French centre, where they received standard treatment. For specific mortality as well as DFS analysis, the ROC curves were greater with the NN models compared to LR model with better sensitivity and specificity. Four predictive factors were retained by both approaches for mortality: clinical size stage, Scarff Bloom Richardson grade, number of invaded nodes, and progesterone receptor. The results enhanced the relevance of the use of NN models in predictive analysis in oncology, which appeared to be more accurate in prediction in this French breast cancer cohort.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.