We present a method for supervised, automatic, and reliable classification of healthy controls, patients with bipolar disorder, and patients with schizophrenia using brain imaging data. The method uses four supervised classification learning machines trained with a stochastic gradient learning rule based on the minimization of KullbackLeibler divergence and an optimal model complexity search through posterior probability estimation. Prior to classification, given the high dimensionality of functional MRI (fMRI) data, a dimension reduction stage comprising two steps is performed: first, a one-sample univariate t-test mean-difference Tscore approach is used to reduce the number of significant discriminative functional activated voxels, and then singular value decomposition is performed to further reduce the dimension of the input patterns to a number comparable to the limited number of subjects available for each of the three classes. Experimental results using functional brain imaging (fMRI) data include receiver operation characteristic curves for the three-way classifier with area under curve values around 0.82, 0.89, and 0.90 for healthy control versus nonhealthy, bipolar disorder versus nonbipolar, and schizophrenia patients versus nonschizophrenia binary problems, respectively. The average three-way correct classification rate (CCR) is in the range of 70\%-72\%, for the test set, remaining close to the estimated Bayesian optimal CCR theoretical upper bound of about 80\%, estimated from the one nearest-neighbor classifier over the same data. {\^A}{\textcopyright} 2010 IEEE.

}, keywords = {Algorithms, area under the curve, article, Artificial Intelligence, Bayesian learning, Bayesian networks, Bayes Theorem, Biological, bipolar disorder, Brain, Case-Control Studies, classification, Classifiers, Computer-Assisted, controlled study, Diseases, functional magnetic resonance imaging, Functional MRI (fMRI), human, Humans, Learning machines, Learning systems, machine learning, Magnetic Resonance Imaging, major clinical study, Models, neuroimaging, Operation characteristic, Optimization, patient coding, receiver operating characteristic, reliability, Reproducibility of Results, ROC Curve, schizophrenia, Signal Processing, Singular value decomposition, Statistical tests, Stochastic models, Student t test}, issn = {00189294}, doi = {10.1109/TBME.2010.2080679}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-78649311169\&partnerID=40\&md5=d3b90f1a3ee4ef209d131ef986e142db}, author = {Juan I. Arribas and V D Calhoun and T Adali} } @article {422, title = {A radius and ulna TW3 bone age assessment system}, journal = {IEEE Transactions on Biomedical Engineering}, volume = {55}, year = {2008}, pages = {1463-1476}, abstract = {An end-to-end system to automate the well-known Tanner - Whitehouse (TW3) clinical procedure to estimate the skeletal age in childhood is proposed. The system comprises the detailed analysis of the two most important bones in TW3: the radius and ulna wrist bones. First, a modified version of an adaptive clustering segmentation algorithm is presented to properly semi-automatically segment the contour of the bones. Second, up to 89 features are defined and extracted from bone contours and gray scale information inside the contour, followed by some well-founded feature selection mathematical criteria, based on the ideas of maximizing the classes{\textquoteright} separability. Third, bone age is estimated with the help of a Generalized Softmax Perceptron (GSP) neural network (NN) that, after supervised learning and optimal complexity estimation via the application of the recently developed Posterior Probability Model Selection (PPMS) algorithm, is able to accurately predict the different development stages in both radius and ulna from which and with the help of the TW3 methodology, we are able to conveniently score and estimate the bone age of a patient in years, in what can be understood as a multiple-class (multiple stages) pattern recognition approach with posterior probability estimation. Finally, numerical results are presented to evaluate the system performance in predicting the bone stages and the final patient bone age over a private hand image database, with the help of the pediatricians and the radiologists expert diagnoses. {\^A}{\textcopyright} 2006 IEEE.

}, keywords = {Age Determination by Skeleton, Aging, algorithm, Algorithms, article, Artificial Intelligence, artificial neural network, Automated, automation, Bone, bone age, Bone age assessment, bone maturation, childhood, Clustering algorithms, Computer-Assisted, Humans, instrumentation, Model selection, Neural networks, Pattern recognition, Radiographic Image Interpretation, radius, Reproducibility of Results, Sensitivity and Specificity, Skeletal maturity, ulna}, issn = {00189294}, doi = {10.1109/TBME.2008.918554}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-42249094547\&partnerID=40\&md5=2cecfea5f75a61b048611f2391b00aed}, author = {Antonio Trist{\'a}n-Vega and Juan I. Arribas} } @article {420, title = {A model selection algorithm for a posteriori probability estimation with neural networks}, journal = {IEEE Transactions on Neural Networks}, volume = {16}, year = {2005}, pages = {799-809}, abstract = {This paper proposes a novel algorithm to jointly determine the structure and the parameters of a posteriori probability model based on neural networks (NNs). It makes use of well-known ideas of pruning, splitting, and merging neural components and takes advantage of the probabilistic interpretation of these components. The algorithm, so called a posteriori probability model selection (PPMS), is applied to an NN architecture called the generalized softmax perceptron (GSP) whose outputs can be understood as probabilities although results shown can be extended to more general network architectures. Learning rules are derived from the application of the expectation-maximization algorithm to the GSP-PPMS structure. Simulation results show the advantages of the proposed algorithm with respect to other schemes. {\^A}{\textcopyright} 2005 IEEE.

}, keywords = {algorithm, Algorithms, article, artificial neural network, Automated, automated pattern recognition, Biological, biological model, Breast Neoplasms, breast tumor, classification, cluster analysis, computer analysis, Computer-Assisted, computer assisted diagnosis, Computer simulation, Computing Methodologies, decision support system, Decision Support Techniques, Diagnosis, Estimation, evaluation, Expectation-maximization, Generalized Softmax Perceptron (GSP), human, Humans, mathematical computing, Mathematical models, methodology, Models, Model selection, Neural networks, Neural Networks (Computer), Numerical Analysis, Objective function, Pattern recognition, Posterior probability, Probability, Statistical, statistical model, statistics, Stochastic Processes}, issn = {10459227}, doi = {10.1109/TNN.2005.849826}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-23044459586\&partnerID=40\&md5=f00e7d86a625cfc466373a2a938276d0}, author = {Juan I. Arribas and Jes{\'u}s Cid-Sueiro} } @conference {417, title = {A radius and ulna skeletal age assessment system}, booktitle = {2005 IEEE Workshop on Machine Learning for Signal Processing}, year = {2005}, address = {Mystic, CT}, abstract = {An end to end system to partially automate the TW3 bone age assessment procedure is proposed. The system comprises the detailed analysis of the two more important bones in TW3: the radius and ulna wrist bones. First, a generalization of K-means algorithm is presented to semi-automatically segment the contour of the bones and thus extract up to 89 features describing shapes and textures from bones. Second, a well-founded feature selection criterion based on the statistical properties of data is used in order to properly choose the most relevant features. Third, bone age is estimated with the help of a Generalized Softmax Perceptron (GSP) Neural Network (NN) whose optimal complexity is estimated via the Posterior Probability Model Selection (PPMS) algorithm. We can then predict the different development stages in both radius and ulna, from which we are able to score and estimate the bone age of a patient in years and finally we compare the NN results with those from the pediatrician expert discrepancies. {\^A}{\textcopyright} 2005 IEEE.

}, keywords = {Algorithms, Bone, Feature extraction, Generalized Softmax Perceptron (GSP), Living systems studies, Neural networks, Probability Model Selection (PPMS), Skeletal age assessment system}, isbn = {0780395174; 9780780395176}, doi = {10.1109/MLSP.2005.1532903}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-33749052083\&partnerID=40\&md5=eefa29ac09f4efa304b613cf07ab8d10}, author = {Antonio Trist{\'a}n-Vega and Juan I. Arribas} } @conference {414, title = {A fully automatic algorithm for contour detection of bones in hand radiographs using active contours}, booktitle = {IEEE International Conference on Image Processing}, year = {2003}, address = {Barcelona}, abstract = {This paper1 presents an algorithm for automatically detecting bone contours from hand radiographs using active contours. Prior knowledge is first used to locate initial contours for the snakes inside each bone of interest. Next, an adaptive snake algorithm is applied so that parameters are properly adjusted for each bone specifically. We introduce a novel truncation technique to prevent the external forces of the snake from pulling the contour outside the bones boundaries, yielding excelent results.

}, keywords = {Active contours, Algorithms, Bone, Cocentric circumferences, Contour measurement, Medical imaging, Object recognition, Radiography}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-0344271749\&partnerID=40\&md5=5fcf06edb482cc1527b2e8d3a940065b}, author = {Rodrigo de Luis-Garc{\'\i}a and Marcos Martin-Fernandez and Juan I. Arribas and Carlos Alberola L{\'o}pez} } @conference {413, title = {Fusing Output Information in Neural Networks: Ensemble Performs Better}, booktitle = {Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, year = {2003}, address = {Cancun}, abstract = {A neural network ensemble is a learning paradigm where a finite number of component neural networks are trained for the same task. Previous research suggests that an ensemble as a whole is often more accurate than any of the single component networks. This paper focuses on the advantages of fusing different nature network architectures, and to determine the appropriate information fusion algorithm in component neural networks by several approaches within hard decision classifiers, when solving a binary pattern recognition problem. We numerically simulated and compared the different fusion approaches in terms of the mean-square error rate in testing data set, over synthetically generated binary Gaussian noisy data, and stated the advantages of fusing the hard outputs of different component networks to make a final hard decision classification. The results of the experiments indicate that neural network ensembles can indeed improve the overall accuracy for classification problems; in all fusion architectures tested, the ensemble correct classification rates are better than those achieved by the individual component networks. Finally we are nowadays comparing the above mentioned hard decision classifiers with new soft decision classifier architectures that make use of the additional continuous type intermediate network soft outputs, fulfilling probability fundamental laws (positive, and add to unity), which can be understood as the a posteriori probabilities of a given pattern to belong to a certain class.

}, keywords = {Algorithms, Backpropagation, Classification (of information), Computer simulation, Decision making, Estimation, Gaussian noise (electronic), Information fusions, Mathematical models, Medical imaging, Model selection, Multilayer neural networks, Neural network ensembles, Pattern recognition, Probability, Probability estimation, Problem solving, Regularization, Statistical methods, Statistical pattern recognition, Vectors}, url = {http://www.scopus.com/inward/record.url?eid=2-s2.0-1542301061\&partnerID=40\&md5=32dbadb3b6ac3c6ae1ea33d89b52c75f}, author = {Y Wu and Juan I. Arribas} }