Shallow and deep learning classifiers in medical image analysis.

An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable of supporting physicians' decision-making. Artificial intelligence encompasses much more than machine learning, which nevertheless is its most cited and used sub-branch in the last decade. Since most clinical problems can be modeled through machine learning classifiers, it is essential to discuss their main elements. This review aims to give primary educational insights on the most accessible and widely employed classifiers in radiology field, distinguishing between "shallow" learning (i.e., traditional machine learning) algorithms, including support vector machines, random forest and XGBoost, and "deep" learning architectures including convolutional neural networks and vision transformers. In addition, the paper outlines the key steps for classifiers training and highlights the differences between the most common algorithms and architectures. Although the choice of an algorithm depends on the task and dataset dealing with, general guidelines for classifier selection are proposed in relation to task analysis, dataset size, explainability requirements, and available computing resources. Considering the enormous interest in these innovative models and architectures, the problem of machine learning algorithms interpretability is finally discussed, providing a future perspective on trustworthy artificial intelligence.Relevance statement The growing synergy between artificial intelligence and medicine fosters predictive models aiding physicians. Machine learning classifiers, from shallow learning to deep learning, are offering crucial insights for the development of clinical decision support systems in healthcare. Explainability is a key feature of models that leads systems toward integration into clinical practice. Key points • Training a shallow classifier requires extracting disease-related features from region of interests (e.g., radiomics).• Deep classifiers implement automatic feature extraction and classification.• The classifier selection is based on data and computational resources availability, task, and explanation needs.

Interpretable Radiomic Signature for Breast Microcalcification Detection and Classification.

Breast microcalcifications are observed in 80% of mammograms, and a notable proportion can lead to invasive tumors. However, diagnosing microcalcifications is a highly complicated and error-prone process due to their diverse sizes, shapes, and subtle variations. In this study, we propose a radiomic signature that effectively differentiates between healthy tissue, benign microcalcifications, and malignant microcalcifications. Radiomic features were extracted from a proprietary dataset, composed of 380 healthy tissue, 136 benign, and 242 malignant microcalcifications ROIs. Subsequently, two distinct signatures were selected to differentiate between healthy tissue and microcalcifications (detection task) and between benign and malignant microcalcifications (classification task). Machine learning models, namely Support Vector Machine, Random Forest, and XGBoost, were employed as classifiers. The shared signature selected for both tasks was then used to train a multi-class model capable of simultaneously classifying healthy, benign, and malignant ROIs. A significant overlap was discovered between the detection and classification signatures. The performance of the models was highly promising, with XGBoost exhibiting an AUC-ROC of 0.830, 0.856, and 0.876 for healthy, benign, and malignant microcalcifications classification, respectively. The intrinsic interpretability of radiomic features, and the use of the Mean Score Decrease method for model introspection, enabled models' clinical validation. In fact, the most important features, namely GLCM Contrast, FO Minimum and FO Entropy, were compared and found important in other studies on breast cancer.

Purification and Biochemical Characterization of Novel Galectin from the Black Poplar Medicinal Mushroom Cyclocybe cylindracea (Agaricomycetes) Strain MEST42 from Algeria.

In the present study, a new galectin designated Cyclocybe cylindracea lectin (CCL) was extracted from the fruiting bodies of the wild black popular mushroom C. cylindracea grown in Algeria. The protein was isolated using sepharose 4B as affinity chromatography matrix, and galactose as elutant. The purified galectin was composed of two subunits of 17.873 kDa each, with a total molecular mass of 35.6 kDa. Its agglutinant activity was impeded by galactose and its derivatives, as well as melibiose. Lactose showed the highest affinity, with a minimal inhibitory concentration of 0.0781 mM. CCL was sensitive to extreme pH conditions, and its binding function decreased when incubated with 10 mM EDTA, and it could be restored by metallic cations such as Ca2+, Mg2+, and Zn2+. CCL agglutinated human red blood cells, without any discernible specificity. Circular dichroism spectra demonstrated that its secondary structure contained ß-sheet as dominant fold. In addition, bioinformatics investigation on their peptide fingerprint obtained after MALDI-TOF/TOF ionization using mascot software confirmed that CCL was not like any previous purified lectin from mushroom: instead, it possessed an amino acid composition with high similarity to that of the putative urea carboxylase of Emericella nidulans (strain FGSC A4/ATCC 38163/CBS 112.46/NRRL 194/M139) with 44% of similarity score.

GOWDL: gene ontology-driven wide and deep learning model for cell typing of scRNA-seq data.

Single-cell RNA-sequencing (scRNA-seq) allows for obtaining genomic and transcriptomic profiles of individual cells. That data make it possible to characterize tissues at the cell level. In this context, one of the main analyses exploiting scRNA-seq data is identifying the cell types within tissue to estimate the quantitative composition of cell populations. Due to the massive amount of available scRNA-seq data, automatic classification approaches for cell typing, based on the most recent deep learning technology, are needed. Here, we present the gene ontology-driven wide and deep learning (GOWDL) model for classifying cell types in several tissues. GOWDL implements a hybrid architecture that considers the functional annotations found in Gene Ontology and the marker genes typical of specific cell types. We performed cross-validation and independent external testing, comparing our algorithm with 12 other state-of-the-art predictors. Classification scores demonstrated that GOWDL reached the best results over five different tissues, except for recall, where we got about 92% versus 97% of the best tool. Finally, we presented a case study on classifying immune cell populations in breast cancer using a hierarchical approach based on GOWDL.

Efficiency of Chitosan Nanocarriers in Vaccinology for Mucosal Immunization.

The mucosal barrier constitutes a huge surface area, close to 40 m2 in humans, located mostly in the respiratory, gastrointestinal and urogenital tracts and ocular cavities. It plays a crucial role in tissue interactions with the microbiome, dietary antigens and other environmental materials. Effective vaccinations to achieve highly protective mucosal immunity are evolving strategies to counteract several serious diseases including tuberculosis, diphtheria, influenzae B, severe acute respiratory syndrome, Human Papilloma Virus infection and Acquired Immune Deficiency Syndrome. Interestingly, one of the reasons behind the rapid spread of severe acute respiratory syndrome coronavirus 2 variants has been the weakness of local immunization at the level of the respiratory mucosa. Mucosal vaccines can outperform parenteral vaccination as they specifically elicit protective mucosal immune responses blocking infection and transmission. In this scenario, chitosan-based nanovaccines are promising adjuvants-carrier systems that rely on the ability of chitosan to cross tight junctions and enhance particle uptake due to chitosan-specific mucoadhesive properties. Indeed, chitosan not only improves the adhesion of antigens to the mucosa promoting their absorption but also shows intrinsic immunostimulant abilities. Furthermore, by finely tuning the colloidal properties of chitosan, it can provide sustained antigen release to strongly activate the humoral defense. In the present review, we agnostically discuss the potential reasons why chitosan-based vaccine carriers, that efficiently elicit strong immune responses in experimental setups and in some pre-clinical/clinical studies, are still poorly considered for therapeutic formulations.

Ln(III) Complexes Embedded in Biocompatible PLGA Nanoparticles as Potential Vis-to-NIR Optical Probes.

In this contribution, we present the spectroscopic study of two NIR emitting hydrophobic heteroleptic (R,R)-YbL1(tta) and (R,R)-NdL1(tta) complexes (with tta = 2-thenoyltrifluoroacetonate and L1 = N,N'-bis(2-(8-hydroxyquinolinate)methylidene)-1,2-(R,R or S,S)-cyclohexanediamine), both in methanol solution and embedded in water dispersible and biocompatible poly lactic-co-glycolic acid (PLGA) nanoparticles. Thanks to their absorption properties in a wide range of wavelengths extending from the UV up to the blue and green visible regions, the emission of these complexes can be effectively sensitized using visible radiation, which is much less harmful to tissues and skin than the UV one. The encapsulation of the two Ln(III)-based complexes in PLGA allows us to preserve their nature, making them stable in water and to test their cytotoxicity on two different cell lines, with the aim of using them in the future as potential bioimaging optical probes.

Effects of Magnetic Nanoparticles on the Functional Activity of Human Monocytes and Dendritic Cells.

The use of nanoparticles in medicine is sometimes hampered by their potential to activate immune cells, eliciting inflammation or allergy. We investigated whether magnetic nanoparticles (MNPs) or biomimetic magnetic nanoparticles (BMNPs) affect relevant activities of human monocytes. We found that the nanoparticles neither elicited the production of pro-inflammatory mediators IL-6 and TNFα by resting monocytes (when BMNP dose < 300 µg/mL) nor enhanced their secretion induced by R848, a molecule engaging virus-recognizing receptors, or bacterial lipopolysaccharide (LPS). MNPs and BMNPs neither induced the generation of reactive oxygen species (ROS), nor affected the ROS production elicited by the NADPH oxidase activator phorbol myristate acetate (PMA) or the fungal derivative ß-glucan. BMNPs, but not MNPs, caused an up-regulation of the maturation markers CD80, CD83, and CD86 in immature monocyte-derived dendritic cells (DCs), whereas both nanoparticles did not affect the LPS-induced expression of these markers. Moreover, the nanoparticles were greedily ingested by monocytes and DCs without altering their viability. Therefore, these nanoparticles are candidates for medical applications because they do not activate pro-inflammatory activities of monocytes. Furthermore, their ability to stimulate DC maturation could be used for the design of vaccines. Moreover, harmlessly engulfed nanoparticles could be vehicles to carry molecules inside the immune cells to regulate the immune response.

Embedding Biomimetic Magnetic Nanoparticles Coupled with Peptide AS-48 into PLGA to Treat Intracellular Pathogens.

Among the strategies employed to overcome the development of multidrug-resistant bacteria, directed chemotherapy combined with local therapies (e.g., magnetic hyperthermia) has gained great interest. A nano-assembly coupling the antimicrobial peptide AS-48 to biomimetic magnetic nanoparticles (AS-48-BMNPs) was demonstrated to have potent bactericidal effects on both Gram-positive and Gram-negative bacteria when the antimicrobial activity of the peptide was combined with magnetic hyperthermia. Nevertheless, intracellular pathogens remain challenging due to the difficulty of the drug reaching the bacterium. Thus, improving the cellular uptake of the nanocarrier is crucial for the success of the treatment. In the present study, we demonstrate the embedding cellular uptake of the original nano-assembly into THP-1, reducing the toxicity of AS-48 toward healthy THP-1 cells. We optimized the design of PLGA[AS-48-BMNPs] in terms of size, colloidal stability, and hyperthermia activity (either magnetic or photothermal). The stability of the nano-formulation at physiological pH values was evaluated by studying the AS-48 release at this pH value. The influence of pH and hyperthermia on the AS-48 release from the nano-formulation was also studied. These results show a slower AS-48 release from PLGA[AS-48-BMNPs] compared to previous nano-formulations, which could make this new nano-formulation suitable for longer extended treatments of intracellular pathogens. PLGA[AS-48-BMNPs] are internalized in THP-1 cells where AS-48 is liberated slowly, which may be useful to treat diseases and prevent infection caused by intracellular pathogens. The treatment will be more efficient combined with hyperthermia or photothermia.

Time-Constrained Node Visit Planning for Collaborative UAV-WSN Distributed Applications.

Unmanned Aerial Vehicles (UAVs) are often studied as tools to perform data collection from Wireless Sensor Networks (WSNs). Path planning is a fundamental aspect of this endeavor. Works in the current literature assume that data are always ready to be retrieved when the UAV passes. This operational model is quite rigid and does not allow for the integration of the UAV as a computational object playing an active role in the network. In fact, the UAV could begin the computation on a first visit and retrieve the data later. Potentially, the UAV could orchestrate the distributed computation to improve its performance, change its parameters, and even upload new applications to the sensor network. In this paper, we analyze a scenario where a UAV plays an active role in the operation of multiple sensor networks by visiting different node clusters to initiate distributed computation and collect the final outcomes. The experimental results validate the effectiveness of the proposed method in optimizing total flight time, Average Age of Information, Average cluster computation end time, and Average data collection time compared to prevalent approaches to UAV path-planning that are adapted to the purpose.

Fisetin: An Integrated Approach to Identify a Strategy Promoting Osteogenesis.

Flavonoids may modulate the bone formation process. Among flavonoids, fisetin is known to counteract tumor growth, osteoarthritis, and rheumatoid arthritis. In addition, fisetin prevents inflammation-induced bone loss. In order to evaluate its favorable use in osteogenesis, we assayed fisetin supplementation in both in vitro and in vivo models and gathered information on nanoparticle-mediated delivery of fisetin in vitro and in a microfluidic system. Real-time RT-PCR, Western blotting, and nanoparticle synthesis were performed to evaluate the effects of fisetin in vitro, in the zebrafish model, and in ex vivo samples. Our results demonstrated that fisetin at 2.5 µM concentration promotes bone formation in vitro and mineralization in the zebrafish model. In addition, we found that fisetin stimulates osteoblast maturation in cell cultures obtained from cleidocranial dysplasia patients. Remarkably, PLGA nanoparticles increased fisetin stability and, consequently, its stimulating effects on RUNX2 and its downstream gene SP7 expression. Therefore, our findings demonstrated the positive effects of fisetin on osteogenesis and suggest that patients affected by skeletal diseases, both of genetic and metabolic origins, may actually benefit from fisetin supplementation.

Near Infared Circularly Polarized Luminescence From Water Stable Organic Nanoparticles Containing a Chiral Yb(III) Complex.

We report the first example of very efficient NIR Circularly Polarized Luminescence (CPL) (around 970 nm) in water, obtained thanks to the combined use of a chiral Yb complex and of poly lactic-co-glycolic acid (PLGA) nanoparticles. [YbL(tta)2 ]CH3 COO (L=N, N'-bis(2-pyridylmethylidene)-1,2-(R,R+S,S) cyclohexanediamine and tta=2-thenoyltrifluoroacetonate) shows good CPL in organic solvents, because the tta ligands efficiently sensitize Yb NIR luminescence and the readily prepared chiral ligand L endows the complex with the necessary dissymmetry. PLGA nanoparticles incorporate the complex and protect the metal ion from the intrusion of solvent molecules, while ensuring biocompatibility, water solubility and stability to the complex. Hydrophilic NIR-CPL optical probes can find applications in the field of NIR-CPL bio-assays.

Characterization of Cytotoxic Lactose Binding Lectin from Sulphur Polypore, Laetiporus sulphureus (Agaricomycetes), from Algeria.

Mushroom lectins have important biological and biomedical applications. Most lectins purified from these organisms exhibit high toxicity in animal cells and toward microbial agents. They are able to induce cell growth inhibition and metabolism by their ability to interact with glyconjugate components (glycoproteins receptors, glycolipids) present in their membrane. After lectins bind to these membrane receptors, they induce cellular signalization chains in which gene expression is regulated and cell death programming (apoptosis) is activated. In this work, a new multimeric lectin was characterized from the rare saprobic edible mushroom, Laetiporus sulphureus strain TMES43, grown in the Algerian forest. Lectin was isolated with ammonium sulfate precipitation followed by affinity chromatography on a Sepharose 4B column, with specific activity of 1204.7 units of hemagglutination activity/mg and 35.55% yield. The protein has a tetrameric structure with a molecular weight of 36 kDa for each subunit, with a total molecular weight of approximately 140 kDa. In addition, a Mascot peptide fingerprint study on a matrix-assisted laser desorption ionization-time of flight tandem fragment showed identity with autophagy-related protein 16 from Meyerozyma guilliermondii (strain ATCC 6260/CBS 566/DSM 6381/JCM 1539/NBRC 10279/NRRL Y-324; Expasy ID: ATG16_PICGU) and no sequence similarity to known mushroom lectins. L. sulphureus hemagglutination activity was reduced by 5 mM of lactose and 10 mM of EDTA incubation and was recovered by metallic cations such as CaCl2, MgCl2, and ZnCl2. L. sulphureus purified lectin had no human ABO group specificity and showed low temperature and alkaline pH stabilities. The MTT preliminary assay showed that L. sulphureus purified lectin induced high cytotoxicity for tumor cells and normal cells.

Enhanced Cytotoxic Effect of TAT-PLGA-Embedded DOXO Carried by Biomimetic Magnetic Nanoparticles upon Combination with Magnetic Hyperthermia and Photothermia.

The synergy between directed chemotherapy and thermal therapy (both magnetic hyperthermia and photothermia) mediated by a nanoassembly composed of functionalized biomimetic magnetic nanoparticles (BMNPs) with the chemotherapeutic drug doxorubicin (DOXO) covered by the polymer poly(lactic-co-glycolic acid) (PLGA), decorated with TAT peptide (here referred to as TAT-PLGA(DOXO-BMNPs)) is explored in the present study. The rationale behind this nanoassembly lies in an optimization of the nanoformulation DOXO-BMNPs, already demonstrated to be more efficient against tumor cells, both in vitro and in vivo, than systemic traditional therapies. By embedding DOXO-BMNPs into PLGA, which is further functionalized with the cell-penetrating TAT peptide, the resulting nanoassembly is able to mediate drug transport (using DOXO as a drug model) and behaves as a hyperthermic agent (induced by an alternating magnetic field (AMF) or by laser irradiation with a laser power density of 2 W/cm2). Our results obtained using the HepG2 cell line show that there is a synergy between chemotherapy and thermal therapy that results in a stronger cytotoxic effect when compared to that caused by the soluble DOXO. This is probably due to the enhanced DOXO release occurring upon the application of the thermal therapy, as well as the induced local temperature rise mediated by BMNPs in the nanoassembly following exposition to AMF or to near-infrared (NIR) laser irradiation. These results represent a proof of concept demonstrating that TAT-PLGA(DOXO-BMNPs) can be used to efficiently combine therapies against tumor cells, which is a step forward in the transition from systemic to local treatments.

Oxyresveratrol-Loaded PLGA Nanoparticles Inhibit Oxygen Free Radical Production by Human Monocytes: Role in Nanoparticle Biocompatibility.

Oxyresveratrol, a polyphenol extracted from the plant Artocarpus lakoocha Roxb, has been reported to be an antioxidant and an oxygen-free radical scavenger. We investigated whether oxyresveratrol affects the generation of superoxide anion (O2-) by human monocytes, which are powerful reactive oxygen species (ROS) producers. We found that oxyresveratrol inhibited the O2- production induced upon stimulation of monocytes with ß-glucan, a well known fungal immune cell activator. We then investigated whether the inclusion of oxyresveratrol into nanoparticles could modulate its effects on O2- release. We synthesized poly(lactic-co-glycolic acid) (PLGA) nanoparticles, and we assessed their effects on monocytes. We found that empty PLGA nanoparticles induced O2- production by resting monocytes and enhanced the formation of this radical in ß-glucan-stimulated monocytes. Interestingly, the insertion of oxyresveratrol into PLGA nanoparticles significantly inhibited the O2- production elicited by unloaded nanoparticles in resting monocytes as well as the synergistic effect of nanoparticles and ß-glucan. Our results indicate that oxyresveratrol is able to inhibit ROS production by activated monocytes, and its inclusion into PLGA nanoparticles mitigates the oxidative effects due to the interaction between these nanoparticles and resting monocytes. Moreover, oxyresveratrol can contrast the synergistic effects of nanoparticles with fungal agents that could be present in the patient tissues. Therefore, oxyresveratrol is a natural compound able to make PLGA nanoparticles more biocompatible.

Oxyresveratrol Inhibits R848-Induced Pro-Inflammatory Mediators Release by Human Dendritic Cells Even When Embedded in PLGA Nanoparticles.

Oxyresveratrol, a stilbene extracted from the plant Artocarpus lakoocha Roxb., has been reported to provide a considerable anti-inflammatory activity. Since the mechanisms of this therapeutic action have been poorly clarified, we investigated whether oxyresveratrol affects the release of the pro-inflammatory cytokines IL-12, IL-6, and TNF-α by human dendritic cells (DCs). We found that oxyresveratrol did not elicit per se the release of these cytokines, but inhibited their secretion induced upon DC stimulation with R848 (Resiquimod), a well-known immune cell activator engaging receptors recognizing RNA viruses. We then investigated whether the inclusion of oxyresveratrol into nanoparticles promoting its ingestion by DCs could favor its effects on cytokine release. For this purpose we synthesized and characterized poly(lactic-co-glycolic acid) (PLGA) nanoparticles, and we assessed their effects on DCs. We found that bare PLGA nanoparticles did not affect cytokine secretion by resting DCs, but increased IL-12, IL-6, and TNF-α secretion by R848-stimulated DCs, an event known as "priming effect". We then loaded PLGA nanoparticles with oxyresveratrol and we observed that oxyresveratrol-bearing particles did not stimulate the cytokine release by resting DCs and inhibited the PLGA-dependent enhancement of IL-12, IL-6, and TNF-α secretion by R848-stimulated DCs. The results herein reported indicate that oxyresveratrol suppresses the cytokine production by activated DCs, thus representing a good anti-inflammatory and immune-suppressive agent. Moreover, its inclusion into PLGA nanoparticles mitigates the pro-inflammatory effects due to cooperation between nanoparticles and R848 in cytokine release. Therefore, oxyresveratrol can be able to contrast the synergistic effects of nanoparticles with microorganisms that could be present in the patient tissues, therefore overcoming a condition unfavorable to the use of some nanoparticles in biological systems.

Knowledge-Based Verification of Concatenative Programming Patterns Inspired by Natural Language for Resource-Constrained Embedded Devices.

We propose a methodology to verify applications developed following programming patterns inspired by natural language that interact with physical environments and run on resource-constrained interconnected devices. Natural language patterns allow for the reduction of intermediate abstraction layers to map physical domain concepts into executable code avoiding the recourse to ontologies, which would need to be shared, kept up to date, and synchronized across a set of devices. Moreover, the computational paradigm we use for effective distributed execution of symbolic code on resource-constrained devices encourages the adoption of such patterns. The methodology is supported by a rule-based system that permits runtime verification of Software Under Test (SUT) on board the target devices through automated oracle and test case generation. Moreover, verification extends from syntactic and semantic checks to the evaluation of the effects of SUT execution on target hardware. Additionally, by exploiting rules tying sensors and actuators to physical quantities, the effects of code execution on the physical environment can be verified. The system is also able to build test code to highlight software issues that may arise during repeated SUT execution on the target hardware.

Deep learning models for bacteria taxonomic classification of metagenomic data.

BACKGROUND: An open challenge in translational bioinformatics is the analysis of sequenced metagenomes from various environmental samples. Of course, several studies demonstrated the 16S ribosomal RNA could be considered as a barcode for bacteria classification at the genus level, but till now it is hard to identify the correct composition of metagenomic data from RNA-seq short-read data. 16S short-read data are generated using two next generation sequencing technologies, i.e. whole genome shotgun (WGS) and amplicon (AMP); typically, the former is filtered to obtain short-reads belonging to a 16S shotgun (SG), whereas the latter take into account only some specific 16S hypervariable regions. The above mentioned two sequencing technologies, SG and AMP, are used alternatively, for this reason in this work we propose a deep learning approach for taxonomic classification of metagenomic data, that can be employed for both of them. RESULTS: To test the proposed pipeline, we simulated both SG and AMP short-reads, from 1000 16S full-length sequences. Then, we adopted a k-mer representation to map sequences as vectors into a numerical space. Finally, we trained two different deep learning architecture, i.e., convolutional neural network (CNN) and deep belief network (DBN), obtaining a trained model for each taxon. We tested our proposed methodology to find the best parameters configuration, and we compared our results against the classification performances provided by a reference classifier for bacteria identification, known as RDP classifier. We outperformed the RDP classifier at each taxonomic level with both architectures. For instance, at the genus level, both CNN and DBN reached 91.3% of accuracy with AMP short-reads, whereas RDP classifier obtained 83.8% with the same data. CONCLUSIONS: In this work, we proposed a 16S short-read sequences classification technique based on k-mer representation and deep learning architecture, in which each taxon (from phylum to genus) generates a classification model. Experimental results confirm the proposed pipeline as a valid approach for classifying bacteria sequences; for this reason, our approach could be integrated into the most common tools for metagenomic analysis. According to obtained results, it can be successfully used for classifying both SG and AMP data.

Antibiotic Resistance Profiling, Analysis of Virulence Aspects and Molecular Genotyping of Staphylococcus aureus Isolated in Sicily, Italy.

Staphylococcus aureus is the major cause of foodborne diseases worldwide. In this retrospective study, 84 S. aureus strains were characterized. The collection comprises 78 strains isolated during 1998 and 2014 from dairy products and tissue samples from livestock bred for dairy production in Sicily. One isolate was obtained from a pet (dog), one from an exotic animal (a circus elephant), and four human isolates were obtained during a severe food poisoning outbreak that occurred in Sicily in 2015. All the strains were characterized by pulsed-field gel electrophoresis (PFGE), for antibiotic resistance and presence of toxin genes. PFGE results showed 10 different pulsotypes, with three relatively frequent and three unique. The antibiotic resistance profiling showed that penicillin G (35.7%) and tetracycline (20.2%) resistance is largely spread. Most isolates contained at least one toxin gene making them a potential threat for public health. Enterotoxin sec gene was observed in 28.6% and seg in 23.8% of the strains, respectively; the human isolates were the only ones to concurrently harbor both seg and sei genes. In addition, 24 isolates were randomly selected and analyzed by multilocus sequence typing. Interestingly, the analysis showed the presence of 12 sequence types (STs), of which 6 were novel. One of them, ST700, was detected in 29% of the isolates and was found to be spread throughout Sicily. ST700 has been present in the island for almost 16 years (1998-2014) and it shows no host preference since it was isolated from different ruminant species. Four human isolates shared both the pulsotype (PT10) and the sequence type (ST9), as well as the virulence genes (seg-sei); this observation suggests that the isolates originated from a single clone, although they were obtained from two different individuals.

Adaptive distributed outlier detection for WSNs.

The paradigm of pervasive computing is gaining more and more attention nowadays, thanks to the possibility of obtaining precise and continuous monitoring. Ease of deployment and adaptivity are typically implemented by adopting autonomous and cooperative sensory devices; however, for such systems to be of any practical use, reliability and fault tolerance must be guaranteed, for instance by detecting corrupted readings amidst the huge amount of gathered sensory data. This paper proposes an adaptive distributed Bayesian approach for detecting outliers in data collected by a wireless sensor network; our algorithm aims at optimizing classification accuracy, time complexity and communication complexity, and also considering externally imposed constraints on such conflicting goals. The performed experimental evaluation showed that our approach is able to improve the considered metrics for latency and energy consumption, with limited impact on classification accuracy.

A knowledge-based decision support system in bioinformatics: an application to protein complex extraction.

BACKGROUND: We introduce a Knowledge-based Decision Support System (KDSS) in order to face the Protein Complex Extraction issue. Using a Knowledge Base (KB) coding the expertise about the proposed scenario, our KDSS is able to suggest both strategies and tools, according to the features of input dataset. Our system provides a navigable workflow for the current experiment and furthermore it offers support in the configuration and running of every processing component of that workflow. This last feature makes our system a crossover between classical DSS and Workflow Management Systems. RESULTS: We briefly present the KDSS' architecture and basic concepts used in the design of the knowledge base and the reasoning component. The system is then tested using a subset of Saccharomyces cerevisiae Protein-Protein interaction dataset. We used this subset because it has been well studied in literature by several research groups in the field of complex extraction: in this way we could easily compare the results obtained through our KDSS with theirs. Our system suggests both a preprocessing and a clustering strategy, and for each of them it proposes and eventually runs suited algorithms. Our system's final results are then composed of a workflow of tasks, that can be reused for other experiments, and the specific numerical results for that particular trial. CONCLUSIONS: The proposed approach, using the KDSS' knowledge base, provides a novel workflow that gives the best results with regard to the other workflows produced by the system. This workflow and its numeric results have been compared with other approaches about PPI network analysis found in literature, offering similar results.