Solving Limited Memory Influence Diagrams

We present a new algorithm for exactly solving decision making problems represented as influence diagrams. We do not require the usual assumptions of no forgetting and regularity; this allows us to solve problems with simultaneous decisions and limited information. The algorithm is empirically shown to outperform a state-of-the-art algorithm on randomly generated problems of up to 150 variables and 10^64 solutions. We show that these problems are NP-hard even if the underlying graph structure of the problem has low treewidth and the variables take on a bounded number of states, and that they admit no provably good approximation if variables can take on an arbitrary number of states.

Decision Making by Credal Nets

Credal nets are probabilistic graphical models which extend Bayesian nets to cope with sets of distributions. This feature makes the model particularly suited for the implementation of classifiers and knowledge-based systems. When working with sets of (instead of single) probability distributions, the identification of the optimal option can be based on different criteria, some of them eventually leading to multiple choices. Yet, most of the inference algorithms for credal nets are designed to compute only the bounds of the posterior probabilities. This prevents some of the existing criteria from being used. To overcome this limitation, we present two simple transformations for credal nets which make it possible to compute decisions based on the maximality and E-admissibility criteria without any modification in the inference algorithms. We also prove that these decision problems have the same complexity of standard inference, being NP^PP-hard for general credal nets and NP-hard for polytrees.

Solving Limited Memory Influence Diagrams

We present a new algorithm for exactly solving decision making problems represented as influence diagrams. We do not require the usual assumptions of no forgetting and regularity; this allows us to solve problems with simultaneous decisions and limited information. The algorithm is empirically shown to outperform a state-of-the-art algorithm on randomly generated problems of up to 150 variables and 10^64 solutions. We show that the problem is NP-hard even if the underlying graph structure of the problem has small treewidth and the variables take on a bounded number of states, but that a fully polynomial time approximation scheme exists for these cases. Moreover, we show that the bound on the number of states is a necessary condition for any efficient approximation scheme.

Generalized Loopy 2U: A New Algorithm for Approximate Inference in Credal Networks

Credal nets generalize Bayesian nets by relaxing the requirement of precision of probabilities. Credal nets are considerably more expressive than Bayesian nets, but this makes belief updating NP-hard even on polytrees. We develop a new efficient algorithm for approximate belief updating in credal nets. The algorithm is based on an important representation result we prove for general credal nets: that any credal net can be equivalently reformulated as a credal net with binary variables; moreover, the transformation, which is considerably more complex than in the Bayesian case, can be implemented in polynomial time. The equivalent binary credal net is updated by L2U, a loopy approximate algorithm for binary credal nets. Thus, we generalize L2U to non-binary credal nets, obtaining an accurate and scalable algorithm for the general case, which is approximate only because of its loopy nature. The accuracy of the inferences is evaluated by empirical tests.

Antimicrobial efficacy of tobramycin polymeric nanoparticles for Pseudomonas aeruginosa infections in cystic fibrosis: formulation, characterisation and functionalisation with dornase alfa (DNase).

Inhaled antibiotics, such as tobramycin, for the treatment of Pseudomonas aeruginosa pulmonary infections are associated with the increase in life expectancy seen in cystic fibrosis (CF) patients over recent years. However, the effectiveness of this aminoglycoside is still limited by its inability to penetrate the thick DNA-rich mucus in the lungs of these patients, leading to low antibiotic exposure to resident bacteria. In this study, we created novel polymeric nanoparticle (NP) delivery vehicles for tobramycin. Using isothermal titration calorimetry, we showed that tobramycin binds with alginate polymer and, by exploiting this interaction, optimised the production of tobramycin alginate/chitosan NPs. It was established that NP antimicrobial activity against P. aeruginosa PA01 was equivalent to unencapsulated tobramycin (minimum inhibitory concentration 0.625 mg/L). Galleria mellonella was employed as an in vivo model for P. aeruginosa infection. Survival rates of 90% were observed following injection of NPs, inferring low NP toxicity. After infection with P. aeruginosa, we showed that a lethal inoculum was effectively cleared by tobramycin NPs in a dose dependent manner. Crucially, a treatment with NPs prior to infection provided a longer window of antibiotic protection, doubling survival rates from 40% with free tobramycin to 80% with NP treatment. Tobramycin NPs were then functionalised with dornase alfa (recombinant human deoxyribonuclease I, DNase), demonstrating DNA degradation and improved NP penetration of CF sputum. Following incubation with CF sputum, tobramycin NPs both with and without DNase functionalisation, exhibited anti-pseudomonal effects. Overall, this work demonstrates the production of effective antimicrobial NPs, which may have clinical utility as mucus-penetrating tobramycin delivery vehicles, combining two widely used CF therapeutics into a single NP formulation. This nano-antibiotic represents a strategy to overcome the mucus barrier, increase local drug concentrations, avoid systemic adverse effects and improve outcomes for pulmonary infections in CF.

Retrieving Regions of Interest for User Exploration

We consider an application scenario where points of interest (PoIs) each have a web presence and where a web user wants to iden- tify a region that contains relevant PoIs that are relevant to a set of keywords, e.g., in preparation for deciding where to go to conve- niently explore the PoIs. Motivated by this, we propose the length- constrained maximum-sum region (LCMSR) query that returns a spatial-network region that is located within a general region of in- terest, that does not exceed a given size constraint, and that best matches query keywords. Such a query maximizes the total weight of the PoIs in it w.r.t. the query keywords. We show that it is NP- hard to answer this query. We develop an approximation algorithm with a (5 + ǫ) approximation ratio utilizing a technique that scales node weights into integers. We also propose a more efficient heuris- tic algorithm and a greedy algorithm. Empirical studies on real data offer detailed insight into the accuracy of the proposed algorithms and show that the proposed algorithms are capable of computingresults efficiently and effectively.

Targeting Siglecs with a sialic acid-decorated nanoparticle abrogates inflammation

Sepsis is the most frequent cause of death in hospitalized patients, and severe sepsis is a leading contributory factor to acute respiratory distress syndrome (ARDS). At present, there is no effective treatment for these conditions, and care is primarily supportive. Murine sialic acid-binding immunoglobulin-like lectin-E (Siglec-E) and its human orthologs Siglec-7 and Siglec-9 are immunomodulatory receptors found predominantly on hematopoietic cells. These receptors are important negative regulators of acute inflammatory responses and are potential targets for the treatment of sepsis and ARDS. We describe a Siglec-targeting platform consisting of poly(lactic-co-glycolic acid) nanoparticles decorated with a natural Siglec ligand, di(α2→8) N-acetylneuraminic acid (α2,8 NANA-NP). This nanoparticle induced enhanced oligomerization of the murine Siglec-E receptor on the surface of macrophages, unlike the free α2,8 NANA ligand. Furthermore, treatment of murine macrophages with these nanoparticles blocked the production of lipopolysaccharide-induced inflammatory cytokines in a Siglec-E-dependent manner. The nanoparticles were also therapeutically beneficial in vivo in both systemic and pulmonary murine models replicating inflammatory features of sepsis and ARDS. Moreover, we confirmed the anti-inflammatory effect of these nanoparticles on human monocytes and macrophages in vitro and in a human ex vivo lung perfusion (EVLP) model of lung injury. We also established that interleukin-10 (IL-10) induced Siglec-E expression and α2,8 NANA-NP further augmented the expression of IL-10. Indeed, the effectiveness of the nanoparticle depended on IL-10. Collectively, these results demonstrated a therapeutic effect of targeting Siglec receptors with a nanoparticle-based platform under inflammatory conditions.

Efficient Processing of Spatial Group Keyword Queries

With the proliferation of geo-positioning and geo-tagging techniques, spatio-textual objects that possess both a geographical location and a textual description are gaining in prevalence, and spatial keyword queries that exploit both location and textual description are gaining in prominence. However, the queries studied so far generally focus on finding individual objects that each satisfy a query rather than finding groups of objects where the objects in a group together satisfy a query.

We define the problem of retrieving a group of spatio-textual objects such that the group's keywords cover the query's keywords and such that the objects are nearest to the query location and have the smallest inter-object distances. Specifically, we study three instantiations of this problem, all of which are NP-hard. We devise exact solutions as well as approximate solutions with provable approximation bounds to the problems. In addition, we solve the problems of retrieving top-k groups of three instantiations, and study a weighted version of the problem that incorporates object weights. We present empirical studies that offer insight into the efficiency of the solutions, as well as the accuracy of the approximate solutions.

Efficient Algorithms for Answering the m-Closest Keywords Query

As an important type of spatial keyword query, the m-closest keywords (mCK) query finds a group of objects such that they cover all query keywords and have the smallest diameter, which is defined as the largest distance between any pair of objects in the group. The query is useful in many applications such as detecting locations of web resources. However, the existing work does not study the intractability of this problem and only provides exact algorithms, which are computationally expensive.

In this paper, we prove that the problem of answering mCK queries is NP-hard. We first devise a greedy algorithm that has an approximation ratio of 2. Then, we observe that an mCK query can be approximately answered by finding the circle with the smallest diameter that encloses a group of objects together covering all query keywords. We prove that the group enclosed in the circle can answer the mCK query with an approximation ratio of 2 over 3. Based on this, we develop an algorithm for finding such a circle exactly, which has a high time complexity. To improve efficiency, we propose another two algorithms that find such a circle approximately, with a ratio of 2 over √3 + ε. Finally, we propose an exact algorithm that utilizes the group found by the 2 over √3 + ε)-approximation algorithm to obtain the optimal group. We conduct extensive experiments using real-life datasets. The experimental results offer insights into both efficiency and accuracy of the proposed approximation algorithms, and the results also demonstrate that our exact algorithm outperforms the best known algorithm by an order of magnitude.

Optimal Route Search with the Coverage of Users' Preferences

The preferences of users are important in route search and planning. For example, when a user plans a trip within a city, their preferences can be expressed as keywords shopping mall, restaurant, and museum, with weights 0.5, 0.4, and 0.1, respectively. The resulting route should best satisfy their weighted preferences. In this paper, we take into account the weighted user preferences in route search, and present a keyword coverage problem, which finds an optimal route from a source location to a target location such that the keyword coverage is optimized and that the budget score satisfies a specified constraint. We prove that this problem is NP-hard. To solve this complex problem, we pro- pose an optimal route search based on an A* variant for which we have defined an admissible heuristic function. The experiments conducted on real-world datasets demonstrate both the efficiency and accu- racy of our proposed algorithms.

A Simple Model to Quantify Radiolytic Production following Electron Emission from Heavy-Atom Nanoparticles Irradiated in Liquid Suspensions

We present a simple model for a component of the radiolytic production of any chemical species due to electron emission from irradiated nanoparticles (NPs) in a liquid environment, provided the expression for the G value for product formation is known and is reasonably well characterized by a linear dependence on beam energy. This model takes nanoparticle size, composition, density and a number of other readily available parameters (such as X-ray and electron attenuation data) as inputs and therefore allows for the ready determination of this contribution. Several approximations are used, thus this model provides an upper limit to the yield of chemical species due to electron emission, rather than a distinct value, and this upper limit is compared with experimental results. After the general model is developed we provide details of its application to the generation of HO(•) through irradiation of gold nanoparticles (AuNPs), a potentially important process in nanoparticle-based enhancement of radiotherapy. This model has been constructed with the intention of making it accessible to other researchers who wish to estimate chemical yields through this process, and is shown to be applicable to NPs of single elements and mixtures. The model can be applied without the need to develop additional skills (such as using a Monte Carlo toolkit), providing a fast and straightforward method of estimating chemical yields. A simple framework for determining the HO(•) yield for different NP sizes at constant NP concentration and initial photon energy is also presented.

The Complexity of MAP Inference in Bayesian Networks Specified Through Logical Languages

We study the computational complexity of finding maximum a posteriori configurations in Bayesian networks whose probabilities are specified by logical formulas. This approach leads to a fine grained study in which local information such as context-sensitive independence and determinism can be considered. It also allows us to characterize more precisely the jump from tractability to NP-hardness and beyond, and to consider the complexity introduced by evidence alone.

Inference in Polytrees with Sets of Probabilities

Inferences in directed acyclic graphs associated with probability intervals and sets of probabilities are NP-hard, even for polytrees. We propose: 1) an improvement on Tessem’s A/R algorithm for inferences on polytrees associated with probability intervals; 2) a new algorithm for approximate inferences based on local search; 3) branch-and-bound algorithms that combine the previous techniques. The first two algorithms produce complementary approximate solutions, while branch-and-bound procedures can generate either exact or approximate solutions. We report improvements on existing techniques for inference with probability sets and intervals, in some cases reducing computational effort by several orders of magnitude.

Microneedles: A New Frontier in Nanomedicine Delivery

This review aims to concisely chart the development of two individual research fields, namely nanomedicines, with specific emphasis on nanoparticles (NP) and microparticles (MP), and microneedle (MN) technologies, which have, in the recent past, been exploited in combinatorial approaches for the efficient delivery of a variety of medicinal agents across the skin. This is an emerging and exciting area of pharmaceutical sciences research within the remit of transdermal drug delivery and as such will undoubtedly continue to grow with the emergence of new formulation and fabrication methodologies for particles and MN. Firstly, the fundamental aspects of skin architecture and structure are outlined, with particular reference to their influence on NP and MP penetration. Following on from this, a variety of different particles are described, as are the diverse range of MN modalities currently under development. The review concludes by highlighting some of the novel delivery systems which have been described in the literature exploiting these two approaches and directs the reader towards emerging uses for nanomedicines in combination with MN.

Transcending epithelial and intracellular biological barriers; a prototype DNA delivery device

Microneedle technology provides the opportunity for the delivery of DNA therapeutics by a non-invasive, patient acceptable route. To deliver DNA successfully requires consideration of both extra and intracellular biological barriers. In this study we present a novel two tier platform; i) a peptide delivery system, termed RALA, that is able to wrap the DNA into nanoparticles, protect the DNA from degradation, enter cells, disrupt endosomes and deliver the DNA to the nucleus of cells ii) a microneedle (MN) patch that will house the nanoparticles within the polymer matrix, breach the skin's stratum corneum barrier and dissolve upon contact with skin interstitial fluid thus releasing the nanoparticles into the skin. Our data demonstrates that the RALA is essential for preventing DNA degradation within the poly(vinylpyrrolidone) (PVP) polymer matrix. In fact the RALA/DNA nanoparticles (NPs) retained functionality when in the MN arrays after 28days and over a range of temperatures. Furthermore the physical strength and structure of the MNs was not compromised when loaded with the NPs. Finally we demonstrated the effectiveness of our MN-NP platform in vitro and in vivo, with systemic gene expression in highly vascularised regions. Taken together this 'smart-system' technology could be applied to a wide range of genetic therapies.