SimClus: an effective algorithm for clustering with a lower bound on similarity

Clustering algorithms generally accept a parameter k from the user, which determines the number of clusters sought. However, in many application domains, like document categorization, social network clustering, and frequent pattern summarization, the proper value of k is difficult to guess. An alternative clustering formulation that does not require k is to impose a lower bound on the similarity between an object and its corresponding cluster representative. Such a formulation chooses exactly one representative for every cluster and minimizes the representative count. It has many additional benefits. For instance, it supports overlapping clusters in a natural way. Moreover, for every cluster, it selects a representative object, which can be effectively used in summarization or semi-supervised classification task. In this work, we propose an algorithm, SimClus, for clustering with lower bound on similarity. It achieves a O(log n) approximation bound on the number of clusters, whereas for the best previous algorithm the bound can be as poor as O(n). Experiments on real and synthetic data sets show that our algorithm produces more than 40% fewer representative objects, yet offers the same or better clustering quality. We also propose a dynamic variant of the algorithm, which can be effectively used in an on-line setting.     

GENOTOXICITY IN CHIRONOMUS KIIENSIS (CHIRONOMIDAE: DIPTERA) AFTER EXPOSURE TO POLLUTED SEDIMENTS FROM RIVERS OF NORTH PENINSULAR MALAYSIA: IMPLICATION FOR ECOTOXICOLOGICAL MONITORING

Rapid industrialization and urbanization has led to increasing input of chemical contaminants into the aquatic environment of Malaysia. Despite the threat civilization poses to the biota, there are still very few relevant studies on ecotoxicological testing of river ecosystems. To overcome this knowledge gap, we examined lethal and genotoxic effects of sediments from different rivers of the northern Malaysia against Chironomus kiiensis, a group well represented in the aquatic fauna of this region. We exposed the larvae to sediments from Selama River (SR), Permatang Rawa River (PRR) and Kilang Ubi River (KUR) at various durations (0, 6, 12, 24 and 48 h). The larval mortality was monitored, whereas DNA damage in survivors' cells was determined using the comet assay. Pollution level indexed by the amounts of heavy metals and other organic contaminants in the sediment showed progressive increases from SR to PRR to KUR. Highly polluted sediments (PRR to KUR) were detrimental to C. kiiensis larvae, most of which did not survive following exposure for long periods. DNA analyses revealed greater damages in nuclei derived from larvae maintained on polluted sediments, in particular, those from KUR. The effects on the genomic material of C. kiiensis larvae occurred in a time‐dependent manner, with damage level increasing as exposure time progressed. Our results highlight the genotoxic properties of polluted sediments. More importantly, this study showed that C. kiiensis larvae could respond to different levels of pollution with respect to exposure time. It is concluded that C. kiiensis larvae is a potential candidate for river ecotoxicological monitoring. Copyright © 2012 John Wiley & Sons, Ltd.     

The future sustainability of water supply in Kuwait

Abstract

The main natural source of water available in Kuwait is the brackish groundwater located in the Kuwait Group and the Dammam aquifers, where the salinity ranges from 4300 to 10200 mg/l and from 2500 to 10000 mg/l, respectively. Limited fresh groundwater resources in Al‐Rawdhatain and Umm Al‐Aish fields, which have a salinity of 359 ‐1737 mg/l, are also available. Most of the groundwater in Kuwait is used for irrigation, domestic purposes, small‐scale industries and for blending with distilled water. Because rainfall is seasonal and less than annual evaporation, the recharge of the groundwater from rainfall is negligible. The objective of this paper is to assess the different water resources in Kuwait in order to make an integrated management plan and to focus on future sustainability. Generally, in the State of Kuwait, there is increasing pressure on the natural water resources because of the exploitation of the aquifers at a rate exceeding that of both the groundwater recharge and population growth (from 1.4 million in 1980 to 2.55 million in 2003). This has lead to the observed decline in groundwater level and to the deterioration of water quality. In the early 1950s, a plan was endorsed in Kuwait to establish seawater desalination plants. By 2005, six desalination plants have been established, with a total capacity of distillation units of 1434.72 ×103 m³/d (315.6 MIGD). However, the gross maximum consumption for the year 2003 reached a value of 1440.17 ×103 m³/d (316.8 MIGD), which is higher than the total capacity of the available distillation units. Given the limitation of conventional water and the shortages of non‐conventional water, along with the increasing population, Kuwait must consider the recycling of wastewater for irrigation, industrial or any other unrestricted non‐potable purpose.     

cTraj: efficient indexing and searching of sequences containing multiple moving objects

Indexing sequences containing multiple moving objects by all features of these objects captured at every clock tick results in huge index structures due to the large number of extracted features in all sampled instances. Thus, the main problems with current systems that index sequences containing multiple moving objects are: huge storage requirements for index structures, slow search time and low accuracy due to lack of representation of the time-varying features of objects. In this paper, a technique called cTraj to address these problems is proposed. For each object in a sequence, cTraj captures the features at sampled instances. Then, it maps the object??s features at each sampled instance from high-dimensional feature space into a point in low-dimensional distance space. The sequence of points of an object in low-dimensional space is considered the time-varying feature trajectory of the object. To reduce storage requirements of an index structure, the sequence of points in each trajectory is represented by a minimum bounding box (MBB). cTraj indexes a sequence by the MBBs of its objects using a spatial access method (SAM), such as an R?tree; thus, greatly reducing storage requirements of the index and speeding up the search time. The cTraj technique does not result in any false dismissal, but the result might contain a few false alarms, which are removed by a two-step refinement process. The experiments show that the proposed cTraj technique produces effective results comparable to those of a sequential method, however much more efficient.     

mfERG_LAB: Software for processing multifocal electroretinography signals

The multifocal electroretinography technique consists of performing sectorized light excitation of the retina and capturing the resulting evoked potential. This provides functional localized information about the state of the retinal neurons. Analysis of multifocal electroretinography signals can be used for diagnosing different types of optic neuropathies (glaucomatous, demyelinating and ischemic ethiology). In order to obtain a reliable diagnosis, it is necessary to apply advanced processing algorithms (morphological, frequency and time-frequency analysis, etc.) to the multifocal electroretinography signal. This paper presents a software application developed in MATLAB(?) (MathWorks Inc., MA) designed to perform advanced multifocal electroretinography signal analysis and classification. This intuitive application, mfERG_LAB, is used to plot the signals, apply various algorithms to them and present the data in an appropriate format. The application's computational power and modular structure make it suitable for use in clinical settings as a powerful and innovative diagnostic tool, as well as in research and teaching settings as a means of assessing new algorithms.     

Fault prediction and the discriminative powers of connectivity-based object-oriented class cohesion metrics

Context

Several metrics have been proposed to measure the extent to which class members are related. Connectivity-based class cohesion metrics measure the degree of connectivity among the class members.

Objective

We propose a new class cohesion metric that has higher discriminative power than any of the existing cohesion metrics. In addition, we empirically compare the connectivity and non-connectivity-based cohesion metrics.

Method

The proposed class cohesion metric is based on counting the number of possible paths in a graph that represents the connectivity pattern of the class members. We theoretically and empirically validate this path connectivity class cohesion (PCCC) metric. The empirical validation compares seven connectivity-based metrics, including PCCC, and 11 non-connectivity-based metrics in terms of discriminative and fault detection powers. The discriminative-power study explores the probability that a cohesion metric will incorrectly determine classes to be cohesively equal when they have different connectivity patterns. The fault detection study investigates whether connectivity-based metrics, including PCCC, better explain the presence of faults from a statistical standpoint in comparison to other non-connectivity-based cohesion metrics, considered individually or in combination.

Results

The theoretical validation demonstrates that PCCC satisfies the key cohesion properties. The results of the empirical studies indicate that, in contrast to other connectivity-based cohesion metrics, PCCC is much better than any comparable cohesion metric in terms of its discriminative power. In addition, the results also indicate that PCCC measures cohesion aspects that are not captured by other metrics, wherein it is considerably better than other connectivity-based metrics but slightly worse than some other non-connectivity-based cohesion metrics in terms of its ability to predict faulty classes.

Conclusion

PCCC is more useful in practice for the applications in which practitioners need to distinguish between the quality of different classes or the quality of different implementations of the same class.     

Successful Management of Valproate Overdose by a Combination of Extracorporeal Therapies

Valproic acid toxicity is not infrequent and is difficult to treat, as there is no specific antidote. In the past, this drug was considered unremovable by extracorporeal methods because of the highly protein bound state (90–95%). However, recent reports suggest that the toxicokinetics of valproate vary considerably from the pharmacokinetics at therapeutic levels, and at higher concentrations protein‐binding sites become saturated. The drug's relatively low molecular weight (144 Da), small volume of distribution (0.13–0.23 L/kg), and saturable protein binding render it potentially amenable to extracorporeal removal, but published experience is scant and there are only a few reported instances in which patients were successfully treated with extracorporeal methods. Here we report the case of a patient with serious valproate toxicity treated with simultaneous “in series” hemodialysis and hemoperfusion followed by continuous veno‐venous hemodiafiltration. A 35‐year‐old homeless male presented to the emergency department after ingesting 120 pills of valproic acid. Initial valproic acid level was 59 µg/mL. Urine drug screen was negative and serum chemistries including LFTs were normal. He was treated with activated charcoal and admitted to the intensive care unit. 4 h later, he developed respiratory failure and became hemodynamically unstable. He was intubated and successfully resuscitated. Subsequently, the valproic acid level was found to be 553 µg/mL and a decision was made to employ extracorporeal methods for drug removal. “In series” hemodialysis and hemoperfusion were done for 4 h and simultaneous blood samples for measurement of valproic acid levels were obtained as blood entered the hemoperfusion column (arterial) and as it exited the hemodialysis membrane (venous). Extraction ratio, whole blood, and plasma clearances were calculated and they compared substantially with the published data. The above measures decreased valproate levels from 572.6 (pre‐dialysis level) to 203.2 µg/mL. This was followed by continuous veno‐venous hemodiafiltration for 18 h, which sustained the drug removal and prevented significant post‐dialytic rebound. This is the first reported instance in which these three methods were administered together successfully in the management of valproic acid toxicity.     

Waveguiding through a two-dimensional metallic photonic crystal

We present a two‐dimensional (2D) finite‐difference time domain simulation of the propagation of light through linear and bent channels in metallic photonic crystals. We took as a starting point the Bozhevolnyi experiment, consisting of the scattering of surface plasmons by a 2D structure of finitely sized periodic gold dots arranged in a triangular lattice of 400‐nm period. We model injection and propagation of light through linear channels of different widths. We also study the behaviour of light in the presence of a 90° bent line defect made in the structure. We show that the confinement depends on the orientation of the input and output line defects. The two cases of ΓM and ΓK orientations are considered and a spectral study for five different wavelengths is carried out.