Classification of human white blood cells using machine learning for stain-free imaging flow cytometry
Imaging flow cytometry (IFC) produces up to 12 different information-rich images of single cells ... more Imaging flow cytometry (IFC) produces up to 12 different information-rich images of single cells at a throughput of 5000 cells per second. Yet often, cell populations are still studied using manual gating, a technique that has several drawbacks. Firstly, it is hard to reproduce. Secondly, it is subjective and biased. And thirdly, it is time-consuming for large experiments. Therefore, it would be advantageous to replace manual gating with an automated process, which could be based on stain-free measurements originating from the brightfield and darkfield image channels. To realise this potential, advanced data analysis methods are required, in particular, machine learning. Previous works have successfully tested this approach on cell cycle phase classification with both a classical machine learning approach based on manually engineered features, and a deep learning approach. In this work, we compare both approaches extensively on the complex problem of white blood cell classification....
Distributed incremental fingerprint identification with reduced database penetration rate using a hierarchical classification based on feature fusion and selection
In the last five years, deep learning methods and particularly Convolutional Neural Networks (CNN... more In the last five years, deep learning methods and particularly Convolutional Neural Networks (CNNs) have exhibited excellent accuracies in many pattern classification problems. Most of the state-of-the-art models apply data-augmentation techniques at the training stage. This paper provides a brief tutorial on data preprocessing and shows its benefits by using the competitive MNIST handwritten digits classification problem. We show and analyze the impact of different preprocessing techniques on the performance of three CNNs, LeNet, Network3 and DropConnect, together with their ensembles. The analyzed transformations are, centering, elastic deformation, translation, rotation and different combinations of them. Our analysis demonstrates that data-preprocessing techniques, such as the combination of elastic deformation and rotation, together with ensembles have a high potential to further improve the state-of-the-art accuracy in MNIST classification.
Nowadays, many disciplines have to deal with big datasets that additionally involve a high number... more Nowadays, many disciplines have to deal with big datasets that additionally involve a high number of features. Feature selection methods aim at eliminating noisy, redundant, or irrelevant features that may deteriorate the classification performance. However, traditional methods lack enough scalability to cope with datasets of millions of instances and extract successful results in a delimited time. This paper presents a feature selection algorithm based on evolutionary computation that uses the MapReduce paradigm to obtain subsets of features from big datasets. The algorithm decomposes the original dataset in blocks of instances to learn from them in the map phase; then, the reduce phase merges the obtained partial results into a final vector of feature weights, which allows a flexible application of the feature selection procedure using a threshold to determine the selected subset of features. The feature selection method is evaluated by using three well-known classifiers (SVM, Log...
Fingerprint recognition has found a reliable application for verification or identification of pe... more Fingerprint recognition has found a reliable application for verification or identification of people in biometrics. Globally, fingerprints can be viewed as valuable traits due to several perceptions observed by the experts; such as the distinctiveness and the permanence on humans and the performance in real applications. Among the main stages of fingerprint recognition, the automated matching phase has received much attention from the early years up to nowadays. This paper is devoted to review and categorize the vast number of fingerprint matching methods proposed in the specialized literature. In particular, we focus on local minutiae-based matching algorithms, which provide good performance with an excellent trade-off between efficacy and efficiency. We identify the main properties and differences of existing methods. Then, we include an experimental evaluation involving the most representative local minutiae-based matching models in both verification and evaluation tasks. The results obtained will be discussed in detail, supporting the description of future directions.
Computational Intelligence (CI) is a field within Artificial Intelligence that has drawn the atte... more Computational Intelligence (CI) is a field within Artificial Intelligence that has drawn the attention of a numerous community of researchers and practitioners. This field is concerned with computational methods inspired on nature and language and targeted for complex real-world problems for which traditional approaches are innefective or infeasible. While a number of different techniques are included within CI, a special effort is made towards their fusion and hybridization looking for systems that gather the stong points of the original components. In particular, CI hosts artificial neural networks, evolutionary algorithms, fuzzy systems and rough sets. Our group is actively involved in developing R packages for different heavily used CI techniques: e.g. RSNNS, Rmalschains, frbs and RoughSets.
Genetic Programming model. However, following the implementation given by the authors the dimensi... more Genetic Programming model. However, following the implementation given by the authors the dimensionality reduction has been intractable (which may be due to the lack of some key details of this algorithm such as initialization of the chromosomes, number of generations, etc.). In the case of [59], even though it only represents OMs of 21 Â 21 blocks, they were codified with gray scale orientations in an image, leading to 11,025 features, which would lead to a high-dimensional problem. Moreover, we were not able to reduce such large dimensionality following the explanations given in the source paper.
This paper reviews the fingerprint classification literature looking at the problem from a double... more This paper reviews the fingerprint classification literature looking at the problem from a double perspective. We first deal with feature extraction methods, including the different models considered for singular point detection and for orientation map extraction. Then, we focus on the different learning models considered to build the classifiers used to label new fingerprints. Taxonomies and classifications for the feature extraction, singular point detection, orientation extraction and learning methods are presented. A critical view of the existing literature have led us to present a discussion on the existing methods and their drawbacks such as difficulty in their reimplementation, lack of details or major differences in their evaluations procedures. On this account, an experimental analysis of the most relevant methods is carried out in the second part of this paper, and a new method based on their combination is presented.
Advances in Intelligent Systems and Computing, 2012
E-Learning is the topic related to the virtualized distance learning by means of electronic commu... more E-Learning is the topic related to the virtualized distance learning by means of electronic communication mechanisms, specifically the Internet. They are based in the use of approaches with diverse functionality (e-mail, Web pages, forums, learning platforms, and so on) as a support of the process of teaching-learning. The Cloud Computing environment rises as a natural platform to provide support to e-Learning systems and also for the implementation of data mining techniques that allow to explore the enormous data bases generated from the former process to extract the inherent knowledge, since it can be dynamically adapted by providing a scalable system for changing necessities along time.
International Journal of Learning Technology, 2014
E-learning is related to virtualised distance learning by means of electronic communication mecha... more E-learning is related to virtualised distance learning by means of electronic communication mechanisms, using its functionality as a support in the process of teaching-learning. When the learning process becomes computerised, educational data mining employs the information generated from the electronic sources to enrich the learning model for academic purposes. To provide support to e-learning systems, cloud computing is set as a natural platform, as it can be dynamically adapted by presenting a scalable system for the changing necessities of the computer resources over time. It also eases the implementation of data mining techniques to work in a distributed scenario, regarding the large databases generated from e-learning. We give an overview of the current state of the structure of cloud computing, and we provide details of the most common infrastructures that have been developed for such a system. We also present some examples of e-learning approaches for cloud computing, and finally, we discuss the suitability of this environment for educational data mining, suggesting the migration of this approach to this computational scenario. He is an active researcher in computational intelligence, where his work covers the whole spectrum from foundations to applications in a number of engineering and scientific areas. His fields of interest are time series analysis and modeling, distributed computational intelligence, high performance computing, cloud computing, data mining, biometrics, and statistical learning theory. He is a member of a number of scientific associations, including IEEE, IEEE Computational Intelligence Society, and EUSFLAT. students, content developers, and experts. The main advantages defined by studying through online tools include flexibility, convenience, ease of access, consistency and repeatability of the proposed tasks.
Uploads
Papers by Daniel Peralta