Machine Learning and Clinical Text. Supporting Health Information Flow

Fluent health information flow is critical for clinical decision-making. However, a considerable part of this information is free-form text and inabilities to utilize it create risks to patient safety and cost-­effective hospital administration. Methods for automated processing of clinical text are emerging. The aim in this doctoral dissertation is to study machine learning and clinical text in order to support health information flow.First, by analyzing the content of authentic patient records, the aim is to specify clinical needs in order to guide the development of machine learning applications.The contributions are a model of the ideal information flow,a model of the problems and challenges in reality, and a road map for the technology development. Second, by developing applications for practical cases,the aim is to concretize ways to support health information flow. Altogether five machine learning applications for three practical cases are described: The first two applications are binary classification and regression related to the practical case of topic labeling and relevance ranking.The third and fourth application are supervised and unsupervised multi-class classification for the practical case of topic segmentation and labeling.These four applications are tested with Finnish intensive care patient records.The fifth application is multi-label classification for the practical task of diagnosis coding. It is tested with English radiology reports.The performance of all these applications is promising. Third, the aim is to study how the quality of machine learning applications can be reliably evaluated.The associations between performance evaluation measures and methods are addressed,and a new hold-out method is introduced.This method contributes not only to processing time but also to the evaluation diversity and quality. The main conclusion is that developing machine learning applications for text requires interdisciplinary, international collaboration. Practical cases are very different, and hence the development must begin from genuine user needs and domain expertise. The technological expertise must cover linguistics,machine learning, and information systems. Finally, the methods must be evaluated both statistically and through authentic user-feedback.

A crisis of social democracy : organized labour and the NDP in an era of neoliberalism

The NDP was founded out of the ashes of the Co-Operative Commonwealth Federation to cooperate with the Canadian Labour Congress to become the 'political arm of organized labour' in Canada. The NDP has long claimed they are the party which represents the policy goals of organized labour in Canada: that the NDP alone will fight for trade union rights, and will fight for Canadian workers. Divergent Paths is an examination of the links between the labour movement and the ND P in an era ofneo-liberalism. Provincial NDP governments have become increasingly neoliberal in their ideological orientation, and have often proved to be no friend to the labour movement when they hold office. The Federal party has never held power, nor have they ever formed the Official Opposition. This thesis charts the progress of the federal NDP as they become more neoliberal from 1988 to 2006, and shows how this trend effects the links between the NDP and labour. Divergent Paths studies each federal election from 1988 to 2006, looking at the interactions between Labour and the NDP during these elections. Elections provide critical junctions to study discourse - party platforms, speeches, and other official documents can be used to examine discourse. Extensive newspaper searches were used to follow campaign events and policy speeches. Studying the party's discourse can be used to determine the ideological orientation of the party itself: the fact that the party's discourse has become neoliberal is a sure sign that the party itself is neoliberal. The NDP continues to drive towards the centre of the political spectrum in an attempt to gain multi-class support. The NDP seems more interested in gaining seats at any cost, rather then promoting the agenda of Labour. As the party attempts to open up to more multi-class support, Labour becomes increasingly marginalised in the party. A rift which arguably started well before the 1988 election was exacerbated during that election; labour encouraged the NDP to campaign solely on the issue of Free Trade, and the NDP did not. The 1993 election saw the rift between the two grow even further as the Federal NDP suffered major blowbacks from the actions of the Ontario NDP. The 1997 and 2000 elections saw the NDP make a deliberate move to the centre of the political spectrum which increasingly marginalised labour. In the 2004 election, Jack Layton made no attempt to move the party back to the left; and in 2006 the link between labour and the NDP was perhaps irreparably damaged when the CAW endorsed the Liberal party in a strategic voting strategy, and the CLC did not endorse the NDP. The NDP is no longer a reliable ally of organized labour. The Canadian labour movement must decide wether the NDP can be 'salvaged' or if the labour movement should end their alliance with the NDP and engage in a new political project.

Scoliosis curve type classification from 3D trunk image

Adolescent idiopathic scoliosis (AIS) is a deformity of the spine manifested by asymmetry and deformities of the external surface of the trunk. Classification of scoliosis deformities according to curve type is used to plan management of scoliosis patients. Currently, scoliosis curve type is determined based on X-ray exam. However, cumulative exposure to X-rays radiation significantly increases the risk for certain cancer. In this paper, we propose a robust system that can classify the scoliosis curve type from non invasive acquisition of 3D trunk surface of the patients. The 3D image of the trunk is divided into patches and local geometric descriptors characterizing the surface of the back are computed from each patch and forming the features. We perform the reduction of the dimensionality by using Principal Component Analysis and 53 components were retained. In this work a multi-class classifier is built with Least-squares support vector machine (LS-SVM) which is a kernel classifier. For this study, a new kernel was designed in order to achieve a robust classifier in comparison with polynomial and Gaussian kernel. The proposed system was validated using data of 103 patients with different scoliosis curve types diagnosed and classified by an orthopedic surgeon from the X-ray images. The average rate of successful classification was 93.3% with a better rate of prediction for the major thoracic and lumbar/thoracolumbar types.

Non invasive classification system of scoliosis curve types using least-squares support vector machines

Objective To determine scoliosis curve types using non invasive surface acquisition, without prior knowledge from X-ray data. Methods Classification of scoliosis deformities according to curve type is used in the clinical management of scoliotic patients. In this work, we propose a robust system that can determine the scoliosis curve type from non invasive acquisition of the 3D back surface of the patients. The 3D image of the surface of the trunk is divided into patches and local geometric descriptors characterizing the back surface are computed from each patch and constitute the features. We reduce the dimensionality by using principal component analysis and retain 53 components using an overlap criterion combined with the total variance in the observed variables. In this work, a multi-class classifier is built with least-squares support vector machines (LS-SVM). The original LS-SVM formulation was modified by weighting the positive and negative samples differently and a new kernel was designed in order to achieve a robust classifier. The proposed system is validated using data from 165 patients with different scoliosis curve types. The results of our non invasive classification were compared with those obtained by an expert using X-ray images. Results The average rate of successful classification was computed using a leave-one-out cross-validation procedure. The overall accuracy of the system was 95%. As for the correct classification rates per class, we obtained 96%, 84% and 97% for the thoracic, double major and lumbar/thoracolumbar curve types, respectively. Conclusion This study shows that it is possible to find a relationship between the internal deformity and the back surface deformity in scoliosis with machine learning methods. The proposed system uses non invasive surface acquisition, which is safe for the patient as it involves no radiation. Also, the design of a specific kernel improved classification performance.

Fully complex-valued radial basis function networks: orthogonal least squares regression and classification

We consider a fully complex-valued radial basis function (RBF) network for regression and classification applications. For regression problems, the locally regularised orthogonal least squares (LROLS) algorithm aided with the D-optimality experimental design, originally derived for constructing parsimonious real-valued RBF models, is extended to the fully complex-valued RBF (CVRBF) network. Like its real-valued counterpart, the proposed algorithm aims to achieve maximised model robustness and sparsity by combining two effective and complementary approaches. The LROLS algorithm alone is capable of producing a very parsimonious model with excellent generalisation performance while the D-optimality design criterion further enhances the model efficiency and robustness. By specifying an appropriate weighting for the D-optimality cost in the combined model selecting criterion, the entire model construction procedure becomes automatic. An example of identifying a complex-valued nonlinear channel is used to illustrate the regression application of the proposed fully CVRBF network. The proposed fully CVRBF network is also applied to four-class classification problems that are typically encountered in communication systems. A complex-valued orthogonal forward selection algorithm based on the multi-class Fisher ratio of class separability measure is derived for constructing sparse CVRBF classifiers that generalise well. The effectiveness of the proposed algorithm is demonstrated using the example of nonlinear beamforming for multiple-antenna aided communication systems that employ complex-valued quadrature phase shift keying modulation scheme. (C) 2007 Elsevier B.V. All rights reserved.

Secondary structure prediction with support vector machines

Motivation: A new method that uses support vector machines (SVMs) to predict protein secondary structure is described and evaluated. The study is designed to develop a reliable prediction method using an alternative technique and to investigate the applicability of SVMs to this type of bioinformatics problem. Methods: Binary SVMs are trained to discriminate between two structural classes. The binary classifiers are combined in several ways to predict multi-class secondary structure. Results: The average three-state prediction accuracy per protein (Q3) is estimated by cross-validation to be 77.07 ± 0.26% with a segment overlap (Sov) score of 73.32 ± 0.39%. The SVM performs similarly to the 'state-of-the-art' PSIPRED prediction method on a non-homologous test set of 121 proteins despite being trained on substantially fewer examples. A simple consensus of the SVM, PSIPRED and PROFsec achieves significantly higher prediction accuracy than the individual methods. Availability: The SVM classifier is available from the authors. Work is in progress to make the method available on-line and to integrate the SVM predictions into the PSIPRED server.

Advanced feature selection methods in multinominal dementia classification from structural MRI data

Recent studies showed that features extracted from brain MRIs can well discriminate Alzheimer’s disease from Mild Cognitive Impairment. This study provides an algorithm that sequentially applies advanced feature selection methods for findings the best subset of features in terms of binary classification accuracy. The classifiers that provided the highest accuracies, have been then used for solving a multi-class problem by the one-versus-one strategy. Although several approaches based on Regions of Interest (ROIs) extraction exist, the prediction power of features has not yet investigated by comparing filter and wrapper techniques. The findings of this work suggest that (i) the IntraCranial Volume (ICV) normalization can lead to overfitting and worst the accuracy prediction of test set and (ii) the combined use of a Random Forest-based filter with a Support Vector Machines-based wrapper, improves accuracy of binary classification.

Complex extreme learning machine applications in terahertz pulsed signals feature sets

This paper presents a novel approach to the automatic classification of very large data sets composed of terahertz pulse transient signals, highlighting their potential use in biochemical, biomedical, pharmaceutical and security applications. Two different types of THz spectra are considered in the classification process. Firstly a binary classification study of poly-A and poly-C ribonucleic acid samples is performed. This is then contrasted with a difficult multi-class classification problem of spectra from six different powder samples that although have fairly indistinguishable features in the optical spectrum, they also possess a few discernable spectral features in the terahertz part of the spectrum. Classification is performed using a complex-valued extreme learning machine algorithm that takes into account features in both the amplitude as well as the phase of the recorded spectra. Classification speed and accuracy are contrasted with that achieved using a support vector machine classifier. The study systematically compares the classifier performance achieved after adopting different Gaussian kernels when separating amplitude and phase signatures. The two signatures are presented as feature vectors for both training and testing purposes. The study confirms the utility of complex-valued extreme learning machine algorithms for classification of the very large data sets generated with current terahertz imaging spectrometers. The classifier can take into consideration heterogeneous layers within an object as would be required within a tomographic setting and is sufficiently robust to detect patterns hidden inside noisy terahertz data sets. The proposed study opens up the opportunity for the establishment of complex-valued extreme learning machine algorithms as new chemometric tools that will assist the wider proliferation of terahertz sensing technology for chemical sensing, quality control, security screening and clinic diagnosis. Furthermore, the proposed algorithm should also be very useful in other applications requiring the classification of very large datasets.

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: The CADDementia challenge

Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimer's disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n = 30) and optionally on data from other sources (e.g., the Alzheimer's Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

HOW FAR do WE GET USING MACHINE LEARNING BLACK-BOXES?

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Analisando o desempenho do ClassAge: um sistema multiagentes para classificação de padrões

The use of multi-agent systems for classification tasks has been proposed in order to overcome some drawbacks of multi-classifier systems and, as a consequence, to improve performance of such systems. As a result, the NeurAge system was proposed. This system is composed by several neural agents which communicate and negotiate a common result for the testing patterns. In the NeurAge system, a negotiation method is very important to the overall performance of the system since the agents need to reach and agreement about a problem when there is a conflict among the agents. This thesis presents an extensive analysis of the NeurAge System where it is used all kind of classifiers. This systems is now named ClassAge System. It is aimed to analyze the reaction of this system to some modifications in its topology and configuration

Evaluation of texture features in hepatic tissue characterization from non-enhanced CT images

Aim of this paper is to evaluate the diagnostic contribution of various types of texture features in discrimination of hepatic tissue in abdominal non-enhanced Computed Tomography (CT) images. Regions of Interest (ROIs) corresponding to the classes: normal liver, cyst, hemangioma, and hepatocellular carcinoma were drawn by an experienced radiologist. For each ROI, five distinct sets of texture features are extracted using First Order Statistics (FOS), Spatial Gray Level Dependence Matrix (SGLDM), Gray Level Difference Method (GLDM), Laws' Texture Energy Measures (TEM), and Fractal Dimension Measurements (FDM). In order to evaluate the ability of the texture features to discriminate the various types of hepatic tissue, each set of texture features, or its reduced version after genetic algorithm based feature selection, was fed to a feed-forward Neural Network (NN) classifier. For each NN, the area under Receiver Operating Characteristic (ROC) curves (Az) was calculated for all one-vs-all discriminations of hepatic tissue. Additionally, the total Az for the multi-class discrimination task was estimated. The results show that features derived from FOS perform better than other texture features (total Az: 0.802+/-0.083) in the discrimination of hepatic tissue.

Bayesian classification with Gaussian processes

We consider the problem of assigning an input vector bfx to one of m classes by predicting P(c|bfx) for c = 1, ldots, m. For a two-class problem, the probability of class 1 given bfx is estimated by s(y(bfx)), where s(y) = 1/(1 + e^-y). A Gaussian process prior is placed on y(bfx), and is combined with the training data to obtain predictions for new bfx points. We provide a Bayesian treatment, integrating over uncertainty in y and in the parameters that control the Gaussian process prior; the necessary integration over y is carried out using Laplace's approximation. The method is generalized to multi-class problems (m >2) using the softmax function. We demonstrate the effectiveness of the method on a number of datasets.

Applying Associative Classifier PGN for Digitised Cultural Heritage Resource Discovery

Resource discovery is one of the key services in digitised cultural heritage collections. It requires intelligent mining in heterogeneous digital content as well as capabilities in large scale performance; this explains the recent advances in classification methods. Associative classifiers are convenient data mining tools used in the field of cultural heritage, by applying their possibilities to taking into account the specific combinations of the attribute values. Usually, the associative classifiers prioritize the support over the confidence. The proposed classifier PGN questions this common approach and focuses on confidence first by retaining only 100% confidence rules. The classification tasks in the field of cultural heritage usually deal with data sets with many class labels. This variety is caused by the richness of accumulated culture during the centuries. Comparisons of classifier PGN with other classifiers, such as OneR, JRip and J48, show the competitiveness of PGN in recognizing multi-class datasets on collections of masterpieces from different West and East European Fine Art authors and movements.