CKLF and IL1B transcript levels at diagnosis are predictive of relapse in children with pre-B-cell acute lymphoblastic leukaemia.

Disease relapse is the greatest cause of treatment failure in paediatric B-cell acute lymphoblastic leukaemia (B-ALL). Current risk stratifications fail to capture all patients at risk of relapse. Herein, we used a machine-learning approach to identify B-ALL blast-secreted factors that are associated with poor survival outcomes. Using this approach, we identified a two-gene expression signature (CKLF and IL1B) that allowed identification of high-risk patients at diagnosis. This two-gene expression signature enhances the predictive value of current at diagnosis or end-of-induction risk stratification suggesting the model can be applied continuously to help guide implementation of risk-adapted therapies.

Co-fuse: a new class discovery analysis tool to identify and prioritize recurrent fusion genes from RNA-sequencing data.

Recurrent oncogenic fusion genes play a critical role in the development of various cancers and diseases and provide, in some cases, excellent therapeutic targets. To date, analysis tools that can identify and compare recurrent fusion genes across multiple samples have not been available to researchers. To address this deficiency, we developed Co-occurrence Fusion (Co-fuse), a new and easy to use software tool that enables biologists to merge RNA-seq information, allowing them to identify recurrent fusion genes, without the need for exhaustive data processing. Notably, Co-fuse is based on pattern mining and statistical analysis which enables the identification of hidden patterns of recurrent fusion genes. In this report, we show that Co-fuse can be used to identify 2 distinct groups within a set of 49 leukemic cell lines based on their recurrent fusion genes: a multiple myeloma (MM) samples-enriched cluster and an acute myeloid leukemia (AML) samples-enriched cluster. Our experimental results further demonstrate that Co-fuse can identify known driver fusion genes (e.g., IGH-MYC, IGH-WHSC1) in MM, when compared to AML samples, indicating the potential of Co-fuse to aid the discovery of yet unknown driver fusion genes through cohort comparisons. Additionally, using a 272 primary glioma sample RNA-seq dataset, Co-fuse was able to validate recurrent fusion genes, further demonstrating the power of this analysis tool to identify recurrent fusion genes. Taken together, Co-fuse is a powerful new analysis tool that can be readily applied to large RNA-seq datasets, and may lead to the discovery of new disease subgroups and potentially new driver genes, for which, targeted therapies could be developed. The Co-fuse R source code is publicly available at https://github.com/sakrapee/co-fuse .

Pedestrian Detection with Spatially Pooled Features and Structured Ensemble Learning.

Many typical applications of object detection operate within a prescribed false-positive range. In this situation the performance of a detector should be assessed on the basis of the area under the ROC curve over that range, rather than over the full curve, as the performance outside the prescribed range is irrelevant. This measure is labelled as the partial area under the ROC curve (pAUC). We propose a novel ensemble learning method which achieves a maximal detection rate at a user-defined range of false positive rates by directly optimizing the partial AUC using structured learning. In addition, in order to achieve high object detection performance, we propose a new approach to extracting low-level visual features based on spatial pooling. Incorporating spatial pooling improves the translational invariance and thus the robustness of the detection process. Experimental results on both synthetic and real-world data sets demonstrate the effectiveness of our approach, and we show that it is possible to train state-of-the-art pedestrian detectors using the proposed structured ensemble learning method with spatially pooled features. The result is the current best reported performance on the Caltech-USA pedestrian detection dataset.

A scalable stagewise approach to large-margin multiclass loss-based boosting.

We present a scalable and effective classification model to train multiclass boosting for multiclass classification problems. A direct formulation of multiclass boosting had been introduced in the past in the sense that it directly maximized the multiclass margin. The major problem of that approach is its high computational complexity during training, which hampers its application to real-world problems. In this paper, we propose a scalable and simple stagewise multiclass boosting method which also directly maximizes the multiclass margin. Our approach offers the following advantages: 1) it is simple and computationally efficient to train. The approach can speed up the training time by more than two orders of magnitude without sacrificing the classification accuracy and 2) like traditional AdaBoost, it is less sensitive to the choice of parameters and empirically demonstrates excellent generalization performance. Experimental results on challenging multiclass machine learning and vision tasks demonstrate that the proposed approach substantially improves the convergence rate and accuracy of the final visual detector at no additional computational cost compared to existing multiclass boosting.

Large-margin Learning of Compact Binary Image Encodings.

The use of high-dimensional features has become a normal practice in many computer vision applications. The large dimension of these features is a limiting factor upon the number of data points which may be effectively stored and processed, however. We address this problem by developing a novel approach to learning a compact binary encoding, which exploits both pair-wise proximity and class-label information on training data set. Exploiting this extra information allows the development of encodings which, although compact, outperform the original high-dimensional features in terms of final classification or retrieval performance. The method is general, in that it is applicable to both non-parametric and parametric learning methods. This generality means that the embedded features are suitable for a wide variety of computer vision tasks, such as image classification and content-based image retrieval. Experimental results demonstrate that the new compact descriptor achieves an accuracy comparable to, and in some cases better than, the visual descriptor in the original space despite being significantly more compact. Moreover, any convex loss function and convex regularization penalty (e.g., `p norm with p 1) can be incorporated into the framework, which provides future flexibility.

RandomBoost: simplified multiclass boosting through randomization.

We propose a novel boosting approach to multiclass classification problems, in which multiple classes are distinguished by a set of random projection matrices in essence. The approach uses random projections to alleviate the proliferation of binary classifiers typically required to perform multiclass classification. The result is a multiclass classifier with a single vector-valued parameter, irrespective of the number of classes involved. Two variants of this approach are proposed. The first method randomly projects the original data into new spaces, while the second method randomly projects the outputs of learned weak classifiers. These methods are not only conceptually simple but also effective and easy to implement. A series of experiments on synthetic, machine learning, and visual recognition data sets demonstrate that our proposed methods could be compared favorably with existing multiclass boosting algorithms in terms of both the convergence rate and classification accuracy.

Incremental training of a detector using online sparse eigendecomposition.

The ability to efficiently and accurately detect objects plays a very crucial role for many computer vision tasks. Recently, offline object detectors have shown a tremendous success. However, one major drawback of offline techniques is that a complete set of training data has to be collected beforehand. In addition, once learned, an offline detector cannot make use of newly arriving data. To alleviate these drawbacks, online learning has been adopted with the following objectives: 1) the technique should be computationally and storage efficient; 2) the updated classifier must maintain its high classification accuracy. In this paper, we propose an effective and efficient framework for learning an adaptive online greedy sparse linear discriminant analysis model. Unlike many existing online boosting detectors, which usually apply exponential or logistic loss, our online algorithm makes use of linear discriminant analysis' learning criterion that not only aims to maximize the class-separation criterion but also incorporates the asymmetrical property of training data distributions. We provide a better alternative for online boosting algorithms in the context of training a visual object detector. We demonstrate the robustness and efficiency of our methods on handwritten digit and face data sets. Our results confirm that object detection tasks benefit significantly when trained in an online manner.

Efficiently learning a detection cascade with sparse eigenvectors.

Real-time object detection has many computer vision applications. Since Viola and Jones proposed the first real-time AdaBoost based face detection system, much effort has been spent on improving the boosting method. In this work, we first show that feature selection methods other than boosting can also be used for training an efficient object detector. In particular, we introduce greedy sparse linear discriminant analysis (GSLDA) for its conceptual simplicity and computational efficiency; and slightly better detection performance is achieved compared with . Moreover, we propose a new technique, termed boosted greedy sparse linear discriminant analysis (BGSLDA), to efficiently train a detection cascade. BGSLDA exploits the sample reweighting property of boosting and the class-separability criterion of GSLDA. Experiments in the domain of highly skewed data distributions (e.g., face detection) demonstrate that classifiers trained with the proposed BGSLDA outperforms AdaBoost and its variants. This finding provides a significant opportunity to argue that AdaBoost and similar approaches are not the only methods that can achieve high detection results for real-time object detection.