Bringing the Algorithms to the Data: Cloud–Based Benchmarking for Medical Image Analysis

One of the main challenges in medical image retrieval is the increasing volume of image data, which render it difficult for domain experts to find relevant information from large data sets. Effective and efficient medical image retrieval systems are required to better manage medical image information. Text‐based image retrieval (TBIR) was very successful in retrieving images with textual descriptions. Several TBIR approaches rely on models based on bag‐of‐words approaches, in which the image retrieval problem turns into one of standard text‐based information retrieval; where the meanings and values of specific medical entities in the text and metadata are ignored in the image representation and retrieval process. However, we believe that TBIR should extract specific medical entities and terms and then exploit these elements to achieve better image retrieval results. Therefore, we propose a novel reranking method based on medical‐image‐dependent features. These features are manually selected by a medical expert from imaging modalities and medical terminology. First, we represent queries and images using only medical‐image‐dependent features such as image modality and image scale. Second, we exploit the defined features in a new reranking method for medical image retrieval. Our motivation is the large influence of image modality in medical image retrieval and its impact on image‐relevance scores. To evaluate our approach, we performed a series of experiments on the medical ImageCLEF data sets from 2009 to 2013. The BM25 model, a language model, and an image‐relevance feedback model are used as baselines to evaluate our approach. The experimental results show that compared to the BM25 model, the proposed model significantly enhances image retrieval performance. We also compared our approach with other state‐of‐the‐art approaches and show that our approach performs comparably to those of the top three runs in the official ImageCLEF competition.

Dynamic Time Warping (DTW), is a simple but efficient technique for matching sequences with rigid deformation. Therefore, it is frequently used for matching shapes in general, and shapes of handwritten words in Document Image Analysis tasks. As DTW is computationally expensive, efficient algorithms for fast computation are crucial. Retrieving images from large scale datasets using DTW, suffers from the constraint of linear searching of all sample in the datasets. Fast approximation algorithms for image retrieval are mostly based on normed spaces where the triangle inequality holds, which is unfortunately not the case with the DTW metric. In this paper we present a novel approach for fast search of handwritten words within large datasets of shapes. The presented approach is based on the Boost-Map [1] algorithm, for embedding the feature space with the DTW measurement to an euclidean space and use the Local Sensitivity Hashing algorithm (LSH) to rank the k-nearest neighbors of a query image. The algorithm, first, processes and embeds objects of the large data sets to a normed space. Fast approximation of k-nearest neighbors using LSH on the embedding space, generates the top kranked samples which are examined using the real DTW distance to give final accurate results. We demonstrate our method on a database of 45; 800 images of word-parts extracted from the IFN/ENIT database [11] and images collected from 51 different writers. Our method achieves a speedup of 4 orders of magnitude over the exact method, at the cost of only a 2:2% reduction in accuracy.

Is power everything? What can we learn from large data sets

Frequent Itemset Mining is one of the most investigated fields of data mining. It is expensive to mine frequent itemsets for a large scale data set. Especially when some data is added into the data set, it is still time-consuming from the scratch to re-compute the complete data set to update the frequent itemsets of the data set. Aiming to improve the performance of frequent itemset mining for large scale and dynamic data set, we propose a new incremental Apriori algorithm based on Spark. It reuses existing results from previous computation to modify the frequent itemsets according to the newly added data, which avoid massive recomputation. The newly proposed algorithm also takes full advantage of distributed resources with the support of Spark. Analysis in theory ensures that the newly proposed algorithm is correct. Experiments on the real-world data set demonstrate that the newly proposed algorithm effectively avoids reduplicated computation and improves the performance of frequent itemset mining with no additional storage overhead.

In conventional distributed machine learning methods, distributed support vector machines (SVM) algorithms are trained over pre-configured intranet/internet environments to find out an optimal classifier. These methods are very complicated and costly for large datasets. Hence, we propose a method that is referred as the Cloud SVM training mechanism (CloudSVM) in a cloud computing environment with MapReduce technique for distributed machine learning applications. Accordingly, (i) SVM algorithm is trained in distributed cloud storage servers that work concurrently; (ii) merge all support vectors in every trained cloud node; and (iii) iterate these two steps until the SVM converges to the optimal classifier function. Single computer is incapable to train SVM algorithm with large scale data sets. The results of this study are important for training of large scale data sets for machine learning applications. We provided that iterative training of splitted data set in cloud computing environment using SVM will converge to a global optimal classifier in finite iteration size.

The massive adoption of smart phones and other mobile devices has generated humongous amount of spatial and spatio-temporal data. The importance of spatial analytics and aggregation is ever-increasing. An important challenge is to support interactive exploration over such data. However, spatial analytics and aggregation using all data points that satisfy a query condition is expensive, especially over large data sets, and could not meet the needs of interactive exploration. To that end, we present novel indexing structures that support spatial online sampling and aggregation on large spatial and spatio-temporal data sets. In spatial online sampling, random samples from the set of spatial (or spatio-temporal) points that satisfy a query condition are generated incrementally in an online fashion. With more and more samples, various spatial analytics and aggregations can be performed in an online, interactive fashion, with estimators that have better accuracy over time. Our design works well for both memory-based and disk-resident data sets, and scales well towards different query and sample sizes. More importantly, our structures are dynamic, hence, they are able to deal with insertions and deletions efficiently. Extensive experiments on large real data sets demonstrate the improvements achieved by our indexing structures compared to other baseline methods.

The value of protein models obtained with automated protein structure prediction depends primarily on their accuracy. Protein model quality assessment is thus critical to select the model that can best answer biologically relevant questions from an ensemble of predictions. However, despite many advances in the field, different methods capture different types of errors, begging the question of which method to use. We introduce TopScore, a meta Model Quality Assessment Program (meta-MQAP) that uses deep neural networks to combine scores from 15 different primary predictors to predict accurate residue-wise and whole-protein error estimates. The predictions on six large independent data sets are highly correlated to superposition-independent errors in the model, achieving a Pearson's Rall2 of 0.93 and 0.78 for whole-protein and residue-wise error predictions, respectively. This is a significant improvement over any of the investigated primary MQAPs, demonstrating that much can be gained by optimally combining different methods and using different and very large data sets.

For many learning algorithms, their learning accuracy will increase as the size of training data increases, forming the well-known learning curve. Usually a learning curve can be fitted by interpolating or extrapolating some points on it with a specified model. The obtained learning curve can then be used to predict the maximum achievable learning accuracy or to estimate the amount of data needed to achieve an expected learning accuracy, both of which will be especially meaningful to data mining on large data sets. Although some models have been proposed to model learning curves, most of them do not test their applicability to large data sets. In this paper, we focus on this issue. We empirically compare six potentially useful models by fitting learning curves of two typical classification algorithms--C4.5 (decision tree) and LOG (logistic discrimination) on eight large UCI benchmark data sets. By using all available data for learning, we fit a full-length learning curve; by using a small portion of the data, we fit a part-length learning curve. The models are then compared in terms of two performances: (1) how well they fit a full-length learning curve, and (2) how well a fitted part-length learning curve can predict learning accuracy at the full length. Experimental results show that the power law (y = a - b * x-c) is the best among the six models in both the performances for the two algorithms and all the data sets. These results support the applicability of learning curves to data mining.

This tutorial describes techniques essential for searching the large multimedia databases that are now common on the Internet. There are up to 10 million songs in commercial music catalogues and over 300 million images stored in online photo services such as Flickr. How can we find the music, videos or images we want? How can we organize such large collections: find duplicates, create links between similar documents, extract and annotate semantic structures from complex audiovisual documents? Conventional methods for handling large data sets, such as hashing, get us part of the way, but those methods may not straightforwardly be used for similarity-based matching and retrieval in audiovisual document collections. On the other hand, several elaborate methods from multimedia retrieval are available for semantic document analysis. Unfortunately, those methods generally do not scale for large data sets. Instead, new classes of algorithms combining the best of the two worlds of large data methods and semantic analysis are needed to handle large multimedia databases. Innovative methods such as locality sensitive hashing, which are based on randomized probes, are the new workhorses. This tutorial covers methods for multimedia retrieval on large document collections. Starting with audio retrieval, we describe both the theory (i.e., randomized algorithms for hashing) and the implementation details (how do you store hash values for millions of songs?). A special focus is on how to combine large data methods with semantically meaningful descriptors in order to facilitate efficient similarity-based retrieval. Besides audio, the tutorial also covers image, 3d motion and video retrieval.

THINGS-data reflects three large-scale neuroimaging and behavioral datasets of object processing in humans, comprising densely sampled functional MRI and magnetoencephalographic recordings, as well as 4.70 million similarity judgments in response to thousands of photographic images for up to 1854 objects.

This paper addresses the problem of association rules mining with large scale data sets using bees behaviors. The bees swarm optimization method have been successfully running on small and medium data size. Nevertheless, when dealing with large benchmark, it is bluntly blocked. Additionally, graphic processor units are massively threaded providing highly intensive computing and very usable by the optimization research community. The parallelization of such method on GPU architecture can be deal large data sets as the case of WebDocs in real time. In this paper, the evaluation process of the solutions is parallelized. Experimental results reveal that the suggested method outperforms the sequential version at the order of ×70 in most data sets, furthermore, the WebDocs benchmark is handled with less than forty hours.Keywordsbees swarm optimizationassociation rule miningparallel algorithmsGPU architecture

Recent advances in unsteady flow visualization

Parallel coordinate plots (PCPs) are commonly used in information visualization to provide insight into multi-variate data. These plots help to spot correlations between variables. PCPs have been successfully applied to unstructured datasets up to a few millions of points. In this paper, we present techniques to enhance the usability of PCPs for the exploration of large, multi-timepoint volumetric data sets, containing tens of millions of points per timestep. The main difficulties that arise when applying PCPs to large numbers of data points are visual clutter and slow performance, making interactive exploration infeasible. Moreover, the spatial context of the volumetric data is usually lost. We describe techniques for preprocessing using data quantization and compression, and for fast GPU-based rendering of PCPs using joint density distributions for each pair of consecutive variables, resulting in a smooth, continuous visualization. Also, fast brushing techniques are proposed for interactive data selection in multiple linked views, including a 3D spatial volume view. These techniques have been successfully applied to three large data sets: Hurricane Isabel (Vis'04 contest), the ionization front instability data set (Vis'08 design contest), and data from a large-eddy simulation of cumulus clouds. With these data, we show how PCPs can be extended to successfully visualize and interactively explore multi-timepoint volumetric datasets with an order of magnitude more data points.

BackgroundMulti-dimensional scaling (MDS) is aimed to represent high dimensional data in a low dimensional space with preservation of the similarities between data points. This reduction in dimensionality is crucial for analyzing and revealing the genuine structure hidden in the data. For noisy data, dimension reduction can effectively reduce the effect of noise on the embedded structure. For large data set, dimension reduction can effectively reduce information retrieval complexity. Thus, MDS techniques are used in many applications of data mining and gene network research. However, although there have been a number of studies that applied MDS techniques to genomics research, the number of analyzed data points was restricted by the high computational complexity of MDS. In general, a non-metric MDS method is faster than a metric MDS, but it does not preserve the true relationships. The computational complexity of most metric MDS methods is over O(N2), so that it is difficult to process a data set of a large number of genes N, such as in the case of whole genome microarray data.ResultsWe developed a new rapid metric MDS method with a low computational complexity, making metric MDS applicable for large data sets. Computer simulation showed that the new method of split-and-combine MDS (SC-MDS) is fast, accurate and efficient. Our empirical studies using microarray data on the yeast cell cycle showed that the performance of K-means in the reduced dimensional space is similar to or slightly better than that of K-means in the original space, but about three times faster to obtain the clustering results. Our clustering results using SC-MDS are more stable than those in the original space. Hence, the proposed SC-MDS is useful for analyzing whole genome data.ConclusionOur new method reduces the computational complexity from O(N3) to O(N) when the dimension of the feature space is far less than the number of genes N, and it successfully reconstructs the low dimensional representation as does the classical MDS. Its performance depends on the grouping method and the minimal number of the intersection points between groups. Feasible methods for grouping methods are suggested; each group must contain both neighboring and far apart data points. Our method can represent high dimensional large data set in a low dimensional space not only efficiently but also effectively.

We investigate the use of biased sampling according to the density of the data set to speed up the operation of general data mining tasks, such as clustering and outlier detection in large multidimensional data sets. In density-biased sampling, the probability that a given point will be included in the sample depends on the local density of the data set. We propose a general technique for density-biased sampling that can factor in user requirements to sample for properties of interest and can be tuned for specific data mining tasks. This allows great flexibility and improved accuracy of the results over simple random sampling. We describe our approach in detail, we analytically evaluate it, and show how it can be optimized for approximate clustering and outlier detection. Finally, we present a thorough experimental evaluation of the proposed method, applying density-biased sampling on real and synthetic data sets, and employing clustering and outlier detection algorithms, thus highlighting the utility of our approach.