Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

PURPOSE: To evaluate the sensitivity and specificity of machine learning classifiers (MLCs) for glaucoma diagnosis using Spectral Domain OCT (SD-OCT) and standard automated perimetry (SAP). METHODS: Observational cross-sectional study. Sixty two glaucoma patients and 48 healthy individuals were included. All patients underwent a complete ophthalmologic examination, achromatic standard automated perimetry (SAP) and retinal nerve fiber layer (RNFL) imaging with SD-OCT (Cirrus HD-OCT; Carl Zeiss Meditec Inc., Dublin, California). Receiver operating characteristic (ROC) curves were obtained for all SD-OCT parameters and global indices of SAP. Subsequently, the following MLCs were tested using parameters from the SD-OCT and SAP: Bagging (BAG), Naive-Bayes (NB), Multilayer Perceptron (MLP), Radial Basis Function (RBF), Random Forest (RAN), Ensemble Selection (ENS), Classification Tree (CTREE), Ada Boost M1(ADA),Support Vector Machine Linear (SVML) and Support Vector Machine Gaussian (SVMG). Areas under the receiver operating characteristic curves (aROC) obtained for isolated SAP and OCT parameters were compared with MLCs using OCT+SAP data. RESULTS: Combining OCT and SAP data, MLCs' aROCs varied from 0.777(CTREE) to 0.946 (RAN).The best OCT+SAP aROC obtained with RAN (0.946) was significantly larger the best single OCT parameter (p<0.05), but was not significantly different from the aROC obtained with the best single SAP parameter (p=0.19). CONCLUSION: Machine learning classifiers trained on OCT and SAP data can successfully discriminate between healthy and glaucomatous eyes. The combination of OCT and SAP measurements improved the diagnostic accuracy compared with OCT data alone.

DECISION TREE CLASSIFIERS FOR STAR/GALAXY SEPARATION

We study the star/galaxy classification efficiency of 13 different decision tree algorithms applied to photometric objects in the Sloan Digital Sky Survey Data Release Seven (SDSS-DR7). Each algorithm is defined by a set of parameters which, when varied, produce different final classification trees. We extensively explore the parameter space of each algorithm, using the set of 884,126 SDSS objects with spectroscopic data as the training set. The efficiency of star-galaxy separation is measured using the completeness function. We find that the Functional Tree algorithm (FT) yields the best results as measured by the mean completeness in two magnitude intervals: 14 <= r <= 21 (85.2%) and r >= 19 (82.1%). We compare the performance of the tree generated with the optimal FT configuration to the classifications provided by the SDSS parametric classifier, 2DPHOT, and Ball et al. We find that our FT classifier is comparable to or better in completeness over the full magnitude range 15 <= r <= 21, with much lower contamination than all but the Ball et al. classifier. At the faintest magnitudes (r > 19), our classifier is the only one that maintains high completeness (> 80%) while simultaneously achieving low contamination (similar to 2.5%). We also examine the SDSS parametric classifier (psfMag - modelMag) to see if the dividing line between stars and galaxies can be adjusted to improve the classifier. We find that currently stars in close pairs are often misclassified as galaxies, and suggest a new cut to improve the classifier. Finally, we apply our FT classifier to separate stars from galaxies in the full set of 69,545,326 SDSS photometric objects in the magnitude range 14 <= r <= 21.

Robot visual localization through local feature fusion: an evaluation of multiple classifiers combination approaches

In the last decade, local image features have been widely used in robot visual localization. In order to assess image similarity, a strategy exploiting these features compares raw descriptors extracted from the current image with those in the models of places. This paper addresses the ensuing step in this process, where a combining function must be used to aggregate results and assign each place a score. Casting the problem in the multiple classifier systems framework, in this paper we compare several candidate combiners with respect to their performance in the visual localization task. For this evaluation, we selected the most popular methods in the class of non-trained combiners, namely the sum rule and product rule. A deeper insight into the potential of these combiners is provided through a discriminativity analysis involving the algebraic rules and two extensions of these methods: the threshold, as well as the weighted modifications. In addition, a voting method, previously used in robot visual localization, is assessed. Furthermore, we address the process of constructing a model of the environment by describing how the model granularity impacts upon performance. All combiners are tested on a visual localization task, carried out on a public dataset. It is experimentally demonstrated that the sum rule extensions globally achieve the best performance, confirming the general agreement on the robustness of this rule in other classification problems. The voting method, whilst competitive with the product rule in its standard form, is shown to be outperformed by its modified versions.

Ensemble classifiers in remote sensing: a comparative analysis

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Discriminative classifiers for speaker recognition

Speaker Recognition, Speaker Verification, Sparse Kernel Logistic Regression, Support Vector Machine

Machine learning with Lipschitz classifiers

Magdeburg, Univ., Fak. für Elektrotechnik und Informationstechnik, Diss., 2010

Textural classification of land cover using support vector machines: an empirical comparison with parametric, non parametric and hybrid classifiers in the Bolivian Amazon

Land cover classification is a key research field in remote sensing and land change science as thematic maps derived from remotely sensed data have become the basis for analyzing many socio-ecological issues. However, land cover classification remains a difficult task and it is especially challenging in heterogeneous tropical landscapes where nonetheless such maps are of great importance. The present study aims to establish an efficient classification approach to accurately map all broad land cover classes in a large, heterogeneous tropical area of Bolivia, as a basis for further studies (e.g., land cover-land use change). Specifically, we compare the performance of parametric (maximum likelihood), non-parametric (k-nearest neighbour and four different support vector machines - SVM), and hybrid classifiers, using both hard and soft (fuzzy) accuracy assessments. In addition, we test whether the inclusion of a textural index (homogeneity) in the classifications improves their performance. We classified Landsat imagery for two dates corresponding to dry and wet seasons and found that non-parametric, and particularly SVM classifiers, outperformed both parametric and hybrid classifiers. We also found that the use of the homogeneity index along with reflectance bands significantly increased the overall accuracy of all the classifications, but particularly of SVM algorithms. We observed that improvements in producer’s and user’s accuracies through the inclusion of the homogeneity index were different depending on land cover classes. Earlygrowth/degraded forests, pastures, grasslands and savanna were the classes most improved, especially with the SVM radial basis function and SVM sigmoid classifiers, though with both classifiers all land cover classes were mapped with producer’s and user’s accuracies of around 90%. Our approach seems very well suited to accurately map land cover in tropical regions, thus having the potential to contribute to conservation initiatives, climate change mitigation schemes such as REDD+, and rural development policies.

Domain Adaptation in remote sensing: increasing the portability of land-cover classifiers

Among the types of remote sensing acquisitions, optical images are certainly one of the most widely relied upon data sources for Earth observation. They provide detailed measurements of the electromagnetic radiation reflected or emitted by each pixel in the scene. Through a process termed supervised land-cover classification, this allows to automatically yet accurately distinguish objects at the surface of our planet. In this respect, when producing a land-cover map of the surveyed area, the availability of training examples representative of each thematic class is crucial for the success of the classification procedure. However, in real applications, due to several constraints on the sample collection process, labeled pixels are usually scarce. When analyzing an image for which those key samples are unavailable, a viable solution consists in resorting to the ground truth data of other previously acquired images. This option is attractive but several factors such as atmospheric, ground and acquisition conditions can cause radiometric differences between the images, hindering therefore the transfer of knowledge from one image to another. The goal of this Thesis is to supply remote sensing image analysts with suitable processing techniques to ensure a robust portability of the classification models across different images. The ultimate purpose is to map the land-cover classes over large spatial and temporal extents with minimal ground information. To overcome, or simply quantify, the observed shifts in the statistical distribution of the spectra of the materials, we study four approaches issued from the field of machine learning. First, we propose a strategy to intelligently sample the image of interest to collect the labels only in correspondence of the most useful pixels. This iterative routine is based on a constant evaluation of the pertinence to the new image of the initial training data actually belonging to a different image. Second, an approach to reduce the radiometric differences among the images by projecting the respective pixels in a common new data space is presented. We analyze a kernel-based feature extraction framework suited for such problems, showing that, after this relative normalization, the cross-image generalization abilities of a classifier are highly increased. Third, we test a new data-driven measure of distance between probability distributions to assess the distortions caused by differences in the acquisition geometry affecting series of multi-angle images. Also, we gauge the portability of classification models through the sequences. In both exercises, the efficacy of classic physically- and statistically-based normalization methods is discussed. Finally, we explore a new family of approaches based on sparse representations of the samples to reciprocally convert the data space of two images. The projection function bridging the images allows a synthesis of new pixels with more similar characteristics ultimately facilitating the land-cover mapping across images.

A Robust Multiple Feature Approach To Endpoint Detection In Car Environment Based On Advanced Classifiers

In this paper we propose an endpoint detection system based on the use of several features extracted from each speech frame, followed by a robust classifier (i.e Adaboost and Bagging of decision trees, and a multilayer perceptron) and a finite state automata (FSA). We present results for four different classifiers. The FSA module consisted of a 4-state decision logic that filtered false alarms and false positives. We compare the use of four different classifiers in this task. The look ahead of the method that we propose was of 7 frames, which are the number of frames that maximized the accuracy of the system. The system was tested with real signals recorded inside a car, with signal to noise ratio that ranged from 6 dB to 30dB. Finally we present experimental results demonstrating that the system yields robust endpoint detection.

Prediction of soil orders with high spatial resolution: response of different classifiers to sampling density

The objective of this work was to evaluate sampling density on the prediction accuracy of soil orders, with high spatial resolution, in a viticultural zone of Serra Gaúcha, Southern Brazil. A digital elevation model (DEM), a cartographic base, a conventional soil map, and the Idrisi software were used. Seven predictor variables were calculated and read along with soil classes in randomly distributed points, with sampling densities of 0.5, 1, 1.5, 2, and 4 points per hectare. Data were used to train a decision tree (Gini) and three artificial neural networks: adaptive resonance theory, fuzzy ARTMap; self‑organizing map, SOM; and multi‑layer perceptron, MLP. Estimated maps were compared with the conventional soil map to calculate omission and commission errors, overall accuracy, and quantity and allocation disagreement. The decision tree was less sensitive to sampling density and had the highest accuracy and consistence. The SOM was the less sensitive and most consistent network. The MLP had a critical minimum and showed high inconsistency, whereas fuzzy ARTMap was more sensitive and less accurate. Results indicate that sampling densities used in conventional soil surveys can serve as a reference to predict soil orders in Serra Gaúcha.

Evaluation of Machine Learning Classifiers for Mobile Network Intrusion Detection Systems

Mobile malwares are increasing with the growing number of Mobile users. Mobile malwares can perform several operations which lead to cybersecurity threats such as, stealing financial or personal information, installing malicious applications, sending premium SMS, creating backdoors, keylogging and crypto-ransomware attacks. Knowing the fact that there are many illegitimate Applications available on the App stores, most of the mobile users remain careless about the security of their Mobile devices and become the potential victim of these threats. Previous studies have shown that not every antivirus is capable of detecting all the threats; due to the fact that Mobile malwares use advance techniques to avoid detection. A Network-based IDS at the operator side will bring an extra layer of security to the subscribers and can detect many advanced threats by analyzing their traffic patterns. Machine Learning(ML) will provide the ability to these systems to detect unknown threats for which signatures are not yet known. This research is focused on the evaluation of Machine Learning classifiers in Network-based Intrusion detection systems for Mobile Networks. In this study, different techniques of Network-based intrusion detection with their advantages, disadvantages and state of the art in Hybrid solutions are discussed. Finally, a ML based NIDS is proposed which will work as a subsystem, to Network-based IDS deployed by Mobile Operators, that can help in detecting unknown threats and reducing false positives. In this research, several ML classifiers were implemented and evaluated. This study is focused on Android-based malwares, as Android is the most popular OS among users, hence most targeted by cyber criminals. Supervised ML algorithms based classifiers were built using the dataset which contained the labeled instances of relevant features. These features were extracted from the traffic generated by samples of several malware families and benign applications. These classifiers were able to detect malicious traffic patterns with the TPR upto 99.6% during Cross-validation test. Also, several experiments were conducted to detect unknown malware traffic and to detect false positives. These classifiers were able to detect unknown threats with the Accuracy of 97.5%. These classifiers could be integrated with current NIDS', which use signatures, statistical or knowledge-based techniques to detect malicious traffic. Technique to integrate the output from ML classifier with traditional NIDS is discussed and proposed for future work.

PERFORMANCE OF DIFFERENT CLASSIFIERS IN SPEECH RECOGNITION

Speech is the most natural means of communication among human beings and speech processing and recognition are intensive areas of research for the last five decades. Since speech recognition is a pattern recognition problem, classification is an important part of any speech recognition system. In this work, a speech recognition system is developed for recognizing speaker independent spoken digits in Malayalam. Voice signals are sampled directly from the microphone. The proposed method is implemented for 1000 speakers uttering 10 digits each. Since the speech signals are affected by background noise, the signals are tuned by removing the noise from it using wavelet denoising method based on Soft Thresholding. Here, the features from the signals are extracted using Discrete Wavelet Transforms (DWT) because they are well suitable for processing non-stationary signals like speech. This is due to their multi- resolutional, multi-scale analysis characteristics. Speech recognition is a multiclass classification problem. So, the feature vector set obtained are classified using three classifiers namely, Artificial Neural Networks (ANN), Support Vector Machines (SVM) and Naive Bayes classifiers which are capable of handling multiclasses. During classification stage, the input feature vector data is trained using information relating to known patterns and then they are tested using the test data set. The performances of all these classifiers are evaluated based on recognition accuracy. All the three methods produced good recognition accuracy. DWT and ANN produced a recognition accuracy of 89%, SVM and DWT combination produced an accuracy of 86.6% and Naive Bayes and DWT combination produced an accuracy of 83.5%. ANN is found to be better among the three methods.

Evolving Classifiers - Evolutionary Algorithms in Data Mining

Data mining means to summarize information from large amounts of raw data. It is one of the key technologies in many areas of economy, science, administration and the internet. In this report we introduce an approach for utilizing evolutionary algorithms to breed fuzzy classifier systems. This approach was exercised as part of a structured procedure by the students Achler, Göb and Voigtmann as contribution to the 2006 Data-Mining-Cup contest, yielding encouragingly positive results.

A Note on the Generalization Performance of Kernel Classifiers with Margin

We present distribution independent bounds on the generalization misclassification performance of a family of kernel classifiers with margin. Support Vector Machine classifiers (SVM) stem out of this class of machines. The bounds are derived through computations of the $V_gamma$ dimension of a family of loss functions where the SVM one belongs to. Bounds that use functions of margin distributions (i.e. functions of the slack variables of SVM) are derived.

Comparing Support Vector Machines with Gaussian Kernels to Radial Basis Function Classifiers

The Support Vector (SV) machine is a novel type of learning machine, based on statistical learning theory, which contains polynomial classifiers, neural networks, and radial basis function (RBF) networks as special cases. In the RBF case, the SV algorithm automatically determines centers, weights and threshold such as to minimize an upper bound on the expected test error. The present study is devoted to an experimental comparison of these machines with a classical approach, where the centers are determined by $k$--means clustering and the weights are found using error backpropagation. We consider three machines, namely a classical RBF machine, an SV machine with Gaussian kernel, and a hybrid system with the centers determined by the SV method and the weights trained by error backpropagation. Our results show that on the US postal service database of handwritten digits, the SV machine achieves the highest test accuracy, followed by the hybrid approach. The SV approach is thus not only theoretically well--founded, but also superior in a practical application.