This chapter introduces the subject of statistical pattern recognition (SPR). It starts by considering how features are defined and emphasizes that the nearest neighbor algorithm achieves error rates comparable with those of an ideal Bayes’ classifier. The concepts of an optimal number of features, representativeness of the training data, and the need to avoid overfitting to the training data are stressed. The chapter shows that methods such as the support vector machine and artificial neural networks are subject to these same training limitations, although each has its advantages. For neural networks, the multilayer perceptron architecture and back-propagation algorithm are described. The chapter distinguishes between supervised and unsupervised learning, demonstrating the advantages of the latter and showing how methods such as clustering and principal components analysis fit into the SPR framework. The chapter also defines the receiver operating characteristic, which allows an optimum balance between false positives and false negatives to be achieved.

Statistical Pattern Recognition

Large, labeled datasets have driven deep learning methods to achieve expert-level performance on a variety of medical imaging tasks. We present CheXpert, a large dataset that contains 224,316 chest radiographs of 65,240 patients. We design a labeler to automatically detect the presence of 14 observations in radiology reports, capturing uncertainties inherent in radiograph interpretation. We investigate different approaches to using the uncertainty labels for training convolutional neural networks that output the probability of these observations given the available frontal and lateral radiographs. On a validation set of 200 chest radiographic studies which were manually annotated by 3 board-certified radiologists, we find that different uncertainty approaches are useful for different pathologies. We then evaluate our best model on a test set composed of 500 chest radiographic studies annotated by a consensus of 5 board-certified radiologists, and compare the performance of our model to that of 3 additional radiologists in the detection of 5 selected pathologies. On Cardiomegaly, Edema, and Pleural Effusion, the model ROC and PR curves lie above all 3 radiologist operating points. We release the dataset to the public as a standard benchmark to evaluate performance of chest radiograph interpretation models.

/pdf/chexpert-a-large-chest-radiograph-dataset-with-uncertainty-4p3sdsnx94.pdf

CheXpert: A large chest radiograph dataset with uncertainty labels and expert comparison

Large, labeled datasets have driven deep learning methods to achieve expert-level performance on a variety of medical imaging tasks. We present CheXpert, a large dataset that contains 224,316 chest radiographs of 65,240 patients. We design a labeler to automatically detect the presence of 14 observations in radiology reports, capturing uncertainties inherent in radiograph interpretation. We investigate different approaches to using the uncertainty labels for training convolutional neural networks that output the probability of these observations given the available frontal and lateral radiographs. On a validation set of 200 chest radiographic studies which were manually annotated by 3 board-certified radiologists, we find that different uncertainty approaches are useful for different pathologies. We then evaluate our best model on a test set composed of 500 chest radiographic studies annotated by a consensus of 5 board-certified radiologists, and compare the performance of our model to that of 3 additional radiologists in the detection of 5 selected pathologies. On Cardiomegaly, Edema, and Pleural Effusion, the model ROC and PR curves lie above all 3 radiologist operating points. We release the dataset to the public as a standard benchmark to evaluate performance of chest radiograph interpretation models. 
The dataset is freely available at this https URL .

CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison

Learning Deconvolution Network for Semantic Segmentation

The $\nu $ -support vector classification has the advantage of using a regularization parameter $\nu $ to control the number of support vectors and margin errors. Recently, a regularization path algorithm for $\nu $ -support vector classification ( $\nu $ -SvcPath) suffers exceptions and singularities in some special cases. In this brief, we first present a new equivalent dual formulation for $\nu $ -SVC and, then, propose a robust $\nu $ -SvcPath, based on lower upper decomposition with partial pivoting. Theoretical analysis and experimental results verify that our proposed robust regularization path algorithm can avoid the exceptions completely, handle the singularities in the key matrix, and fit the entire solution path in a finite number of steps. Experimental results also show that our proposed algorithm fits the entire solution path with fewer steps and less running time than original one does.

A Robust Regularization Path Algorithm for $\nu $ -Support Vector Classification

Among algorithms for comparing the areas under two or more correlated receiver operating characteristic (ROC) curves, DeLong's algorithm is perhaps the most widely used one due to its simplicity of implementation in practice. Unfortunately, however, the time complexity of DeLong's algorithm is of quadratic order (the product of sample sizes), thus making it time-consuming and impractical when the sample sizes are large. Based on an equivalent relationship between the Heaviside function and mid-ranks of samples, we improve DeLong's algorithm by reducing the order of time complexity from quadratic down to linearithmic (the product of sample size and its logarithm). Monte Carlo simulations verify the computational efficiency of our algorithmic findings in this work.

Fast Implementation of DeLong’s Algorithm for Comparing the Areas Under Correlated Receiver Operating Characteristic Curves

A comparative analysis of Spearman's rho and Kendall's tau in normal and contaminated normal models

The performance of the existing sparse Bayesian learning (SBL) methods for off-grid direction-of-arrival (DOA) estimation is dependent on the tradeoff between the accuracy and the computational workload. To speed up the off-grid SBL method while remain a reasonable accuracy, this letter describes a computationally efficient root SBL method for off-grid DOA estimation, which adopts a coarse grid and considers the sampled locations in the coarse grid as the adjustable parameters. We utilize an expectation–maximization algorithm to iteratively refine this coarse grid and illustrate that each updated grid point can be simply achieved by the root of a certain polynomial. Simulation results demonstrate that the computational complexity is significantly reduced, and the modeling error can be almost eliminated.

Root Sparse Bayesian Learning for Off-Grid DOA Estimation

The effect of mutual coupling between array elements is known to substantially degrade direction-of-arrival (DOA) estimation performance. Rank-reduction (RARE) methods can eliminate the mutual coupling effect without loss of array aperture. However, they might yield serious ambiguous DOA estimates in some scenarios. This letter proposes a modified RARE algorithm for DOA estimation, which overcomes the problem of ambiguous estimation by recursively building up RARE matrices. Moreover, due to the utilization of an additional unique structure of mutual coupling coefficients, the new method can give a better DOA estimation performance. Simulation results verify the efficiency of the proposed method.

Weichao Xu

Papers

Fast Implementation of DeLong’s Algorithm for Comparing the Areas Under Correlated Receiver Operating Characteristic Curves

A comparative analysis of Spearman's rho and Kendall's tau in normal and contaminated normal models

Root Sparse Bayesian Learning for Off-Grid DOA Estimation

A Recursive RARE Algorithm for DOA Estimation With Unknown Mutual Coupling

Real-valued DOA estimation for uniform linear array with unknown mutual coupling