Multi-class learning is an important task in Data Science. One of the ways to achieve good performance on this task is to use Error Correcting Output Codes (ECOC), which is a powerful ensemble learning method that transforms a multi-class problem into a series of binary classifiers which it uses indirectly to learn the original multi-class problem. A crucial component of ECOC is the design of the coding matrix, which determines which binary problems should be combined to achieve multi-class classification. There are two general ways of designing the coding matrix. One is rooted in information theory while the other is data driven. In this work, we investigate the data-driven approach which was previously shown to bear greater promise and propose a better search through the coding-matrix space, keeping in mind the tradeoff between efficiency and effectiveness, as well as considerations about class-imbalance issues in the underlying binary problems. After consolidating our hy! potheses with a study on artificial domains, we propose the Unsupervised Deep Search Algorithm (UDS) coupled with re-sampling, to address both concerns. Our results on real world domains show that our method outperforms traditional multi-class learning methods.

BACKGROUND: The subcellular distribution of synapses is fundamentally important for the assembly, function, and plasticity of the nervous system. Automated and effective quantification tools are a prerequisite to large-scale studies of the molecular mechanisms of subcellular synapse distribution. Common practices for synapse quantification in neuroscience labs remain largely manual or semi-manual. This is mainly due to computational challenges in automatic quantification of synapses, including large volume, high dimensions and staining artifacts. In the case of confocal imaging, optical limit and xy-z resolution disparity also require special considerations to achieve the necessary robustness. RESULTS: A novel algorithm is presented in the paper for learning-guided automatic recognition and quantification of synaptic markers in 3D confocal images. The method developed a discriminative model based on 3D feature descriptors that detected the centers of synaptic markers. It made use of adaptive thresholding and multi-channel co-localization to improve the robustness. The detected markers then guided the splitting of synapse clumps, which further improved the precision and recall of the detected synapses. Algorithms were tested on lobula plate tangential cells (LPTCs) in the brain of Drosophila melanogaster, for GABAergic synaptic markers on axon terminals as well as dendrites. CONCLUSIONS: The presented method was able to overcome the staining artifacts and the fuzzy boundaries of synapse clumps in 3D confocal image, and automatically quantify synaptic markers in a complex neuron such as LPTC. Comparison with some existing tools used in automatic 3D synapse quantification also proved the effectiveness of the proposed method.

BACKGROUND: Pattern recognition algorithms are useful in bioimage informatics applications such as quantifying cellular and subcellular objects, annotating gene expressions, and classifying phenotypes. To provide effective and efficient image classification and annotation for the ever-increasing microscopic images, it is desirable to have tools that can combine and compare various algorithms, and build customizable solution for different biological problems. However, current tools often offer a limited solution in generating user-friendly and extensible tools for annotating higher dimensional images that correspond to multiple complicated categories. RESULTS: We develop the BIOimage Classification and Annotation Tool (BIOCAT). It is able to apply pattern recognition algorithms to two- and three-dimensional biological image sets as well as regions of interest (ROIs) in individual images for automatic classification and annotation. We also propose a 3D anisotropic wavelet feature extractor for extracting textural features from 3D images with xy-z resolution disparity. The extractor is one of the about 20 built-in algorithms of feature extractors, selectors and classifiers in BIOCAT. The algorithms are modularized so that they can be "chained" in a customizable way to form adaptive solution for various problems, and the plugin-based extensibility gives the tool an open architecture to incorporate future algorithms. We have applied BIOCAT to classification and annotation of images and ROIs of different properties with applications in cell biology and neuroscience. CONCLUSIONS: BIOCAT provides a user-friendly, portable platform for pattern recognition based biological image classification of two- and three- dimensional images and ROIs. We show, via diverse case studies, that different algorithms and their combinations have different suitability for various problems. The customizability of BIOCAT is thus expected to be useful for providing effective and efficient solutions for a variety of biological problems involving image classification and annotation. We also demonstrate the effectiveness of 3D anisotropic wavelet in classifying both 3D image sets and ROIs.

Brain computer interface (BCI) has gained a lot of attention recently, as a means to detect individuals' intents using brain signals such as electroencephalographic (EEG) for control of machines. In order to achieve the possible use of BCI in stroke rehabilitation, computational intelligent algorithms are important for reliable separation of shoulder versus elbow movement intentions. Efforts have been made on developing data processing and classification algorithm for such task. Differently, this paper investigates the optimal use of electrodes and signal channels, which is formulated as a data-driven feature selection problem. 163 EEG electrodes are used to collect scalp recordings to predict shoulder abduction and elbow flexion intentions in healthy and stroke subjects. We combine the support vector channel selection with a time-frequency synthesized classification algorithm and examine the performances of using different subsets of channel inputs. Preliminary results show that 1) a reduced number of electrodes can be used to achieve the same or better performance than using the full set of signal channels; 2) besides the fact that the accuracy on able-bodied subjects is expectedly higher than the stroke subject, the stroke subject tends to need more electrodes to achieve the best performance; 3) visualization of spatial distribution of channel rankings shows reasonable connection with functional motor cortex areas

Abstract Motivation: Gene expression patterns obtained by in situ mRNA hybridization provide important information about different genes during Drosophila embryogenesis. So far, annotations of these images are done by manually assigning a subset of anatomy ontology terms to an image. This time-consuming process depends heavily on the consistency of experts. Results: We develop a system to automatically annotate a fruitfly's embryonic tissue in which a gene has expression. We formulate the task as an image pattern recognition problem. For a new fly embryo image, our system answers two questions: (1) Which stage range does an image belong to? (2) Which annotations should be assigned to an image? We propose to identify the wavelet embryo features by multi-resolution 2D wavelet discrete transform, followed by min-redundancy max-relevance feature selection, which yields optimal distinguishing features for an annotation. We then construct a series of parallel bi-class predictors to solve the multi-objective annotation problem since each image may correspond to multiple annotations. Supplementary information: The complete annotation prediction results are available at: http://www.cs.niu.edu/~jzhou/papers/fruitfly and http://research.janelia.org/peng/proj/fly_embryo_annotation/. The datasets used in experiments will be available upon request to the correspondence author. Contact: [email protected] and [email protected]

MOTIVATION: Gene expression patterns obtained by in situ mRNA hybridization provide important information about different genes during Drosophila embryogenesis. So far, annotations of these images are done by manually assigning a subset of anatomy ontology terms to an image. This time-consuming process depends heavily on the consistency of experts. RESULTS: We develop a system to automatically annotate a fruitfly's embryonic tissue in which a gene has expression. We formulate the task as an image pattern recognition problem. For a new fly embryo image, our system answers two questions: (1) Which stage range does an image belong to? (2) Which annotations should be assigned to an image? We propose to identify the wavelet embryo features by multi-resolution 2D wavelet discrete transform, followed by min-redundancy max-relevance feature selection, which yields optimal distinguishing features for an annotation. We then construct a series of parallel bi-class predictors to solve the multi-objective annotation problem since each image may correspond to multiple annotations. SUPPLEMENTARY INFORMATION: The complete annotation prediction results are available at: http://www.cs.niu.edu/~jzhou/papers/fruitfly and http://research.janelia.org/peng/proj/fly_embryo_annotation/. The datasets used in experiments will be available upon request to the correspondence author.

This paper describes a new approach to recognize touching numeral strings. Currently most methods for numeral string recognition require segmenting the string image into separate numerals. As a result, the recognition system heavily depends on the reliability of the segmentation module. This study explores the holistic strategy directly on the string images without segmentation. It builds the novel classifier by combining binary classifiers based on data-driven error correcting output coding (DECOC). The dimensions of input images are reduced using principal components analysis. Support vector machines are used as base learners. Experiments on NIST SD19 touching numeral pairs confirm that DECOC can achieve favorable performance compared with other multi-class holistic classifiers. The method provides the flexibility of controlling the computational complexity versus accuracy. We also discuss an implementation suitable for distributing computing by decomposing the ensemble into subtasks.

Premature ventricular contractions (PVCs) are ectopic heart beats originating from ventricular area. It is a common form of heart arrhythmia. Electrocardiogram (ECG) recordings have been widely used to assist cardiologists to diagnose the problem. In this paper, we study the automatic detection of PVC using a fuzzy artificial neural network named Quantum Neural Network (QNN). With the quantum neurons in the network, trained QNN can model the levels of uncertainty arising from complex classification problems. This fuzzy feature is expected to enhance the reliability of the algorithm, which is critical for the applications in the biomedical domain. Experiments were conducted on ECG records in the MIT-BIH Arrhythmia Database. Results showed consistently higher or same reliability of QNN on all the available records compared to the backpropagation network. QNN, however, has a relatively higher resource requirement for training.

The paper describes a high performance offline system for recognizing hand printed numerals. An innovative verification module is applied which drastically improves the recognition rate. The approaches used in the modules are described. The importance of the verification module is analysed in detail. A practical automatic form reading system TOCR V1.0 was developed based on the algorithms. The system was put into practical use in several provinces of China for statistical analysis of Revenue China. Test results are given based on: 1) data collected when the system was used in China, as well as 2) the CENPARMI database.