Shallow Dialogue Processing Using Machine Learning Algorithms (or not)

This paper presents a shallow dialogue analysis model, aimed at human-human dialogues in the context of staff or business meetings. Four components of the model are defined, and several machine learning techniques are used to extract features from dialogue transcripts: maximum entropy classifiers for dialogue acts, latent semantic analysis for topic segmentation, or decision tree classifiers for discourse markers. A rule-based approach is proposed for solving cross-modal references to meeting documents. The methods are trained and evaluated thanks to a common data set and annotation format. The integration of the components into an automated shallow dialogue parser opens the way to multimodal meeting processing and retrieval applications.

Radiation metabolomics. 3. Biomarker discovery in the urine of gamma-irradiated rats using a simplified metabolomics protocol of gas chromatography-mass spectrometry combined with random forests machine learning algorithm

Abstract Radiation metabolomics employing mass spectral technologies represents a plausible means of high-throughput minimally invasive radiation biodosimetry. A simplified metabolomics protocol is described that employs ubiquitous gas chromatography-mass spectrometry and open source software including random forests machine learning algorithm to uncover latent biomarkers of 3 Gy gamma radiation in rats. Urine was collected from six male Wistar rats and six sham-irradiated controls for 7 days, 4 prior to irradiation and 3 after irradiation. Water and food consumption, urine volume, body weight, and sodium, potassium, calcium, chloride, phosphate and urea excretion showed major effects from exposure to gamma radiation. The metabolomics protocol uncovered several urinary metabolites that were significantly up-regulated (glyoxylate, threonate, thymine, uracil, p-cresol) and down-regulated (citrate, 2-oxoglutarate, adipate, pimelate, suberate, azelaate) as a result of radiation exposure. Thymine and uracil were shown to derive largely from thymidine and 2'-deoxyuridine, which are known radiation biomarkers in the mouse. The radiation metabolomic phenotype in rats appeared to derive from oxidative stress and effects on kidney function. Gas chromatography-mass spectrometry is a promising platform on which to develop the field of radiation metabolomics further and to assist in the design of instrumentation for use in detecting biological consequences of environmental radiation release.

Combining Wireless Sensor Networks and Machine Learning for Flash Flood Nowcasting

This paper addresses an investigation with machine learning (ML) classification techniques to assist in the problem of flash flood now casting. We have been attempting to build a Wireless Sensor Network (WSN) to collect measurements from a river located in an urban area. The machine learning classification methods were investigated with the aim of allowing flash flood now casting, which in turn allows the WSN to give alerts to the local population. We have evaluated several types of ML taking account of the different now casting stages (i.e. Number of future time steps to forecast). We have also evaluated different data representation to be used as input of the ML techniques. The results show that different data representation can lead to results significantly better for different stages of now casting.

Fast Prediction of Femoral Biomechanics Using Supervised Machine Learning and Statistical Shape Modeling

Finite element (FE) analysis is an important computational tool in biomechanics. However, its adoption into clinical practice has been hampered by its computational complexity and required high technical competences for clinicians. In this paper we propose a supervised learning approach to predict the outcome of the FE analysis. We demonstrate our approach on clinical CT and X-ray femur images for FE predictions ( FEP), with features extracted, respectively, from a statistical shape model and from 2D-based morphometric and density information. Using leave-one-out experiments and sensitivity analysis, comprising a database of 89 clinical cases, our method is capable of predicting the distribution of stress values for a walking loading condition with an average correlation coefficient of 0.984 and 0.976, for CT and X-ray images, respectively. These findings suggest that supervised learning approaches have the potential to leverage the clinical integration of mechanical simulations for the treatment of musculoskeletal conditions.

Soft tissue grafting to improve the attached mucosa at dental implants: A review of the literature and proposal of a decision tree.

BACKGROUND Scientific data and clinical observations appear to indicate that an adequate width of attached mucosa may facilitate oral hygiene procedures thus preventing peri-implant inflammation and tissue breakdown (eg, biologic complications). Consequently, in order to avoid biologic complications and improve long-term prognosis, soft tissue conditions should be carefully evaluated when implant therapy is planned. At present the necessity and time-point for soft tissue grafting (eg, prior to or during implant placement or after healing) is still controversially discussed while clinical recommendations are vague. OBJECTIVES To provide a review of the literature on the role of attached mucosa to maintain periimplant health, and to propose a decision tree which may help the clinician to select the appropriate surgical technique for increasing the width of attached mucosa. RESULTS The available data indicate that ideally, soft tissue conditions should be optimized by various grafting procedures either before or during implant placement or as part of stage-two surgery. In cases, where, despite insufficient peri-implant soft tissue condition (ie, lack of attached mucosa or movements caused by buccal frena), implants have been uncovered and/or loaded, or in cases where biologic complications are already present (eg, mucositis, peri-implantitis), the treatment appears to be more difficult and less predictable. CONCLUSION Soft tissue grafting may be important to prevent peri-implant tissue breakdown and should be considered when dental implants are placed. The presented decision tree may help the clinician to select the appropriate grafting technique.

Metabolomics reveals attenuation of the SLC6A20 kidney transporter in nonhuman primate and mouse models of type 2 diabetes mellitus

To enhance understanding of the metabolic indicators of type 2 diabetes mellitus (T2DM) disease pathogenesis and progression, the urinary metabolomes of well characterized rhesus macaques (normal or spontaneously and naturally diabetic) were examined. High-resolution ultra-performance liquid chromatography coupled with the accurate mass determination of time-of-flight mass spectrometry was used to analyze spot urine samples from normal (n = 10) and T2DM (n = 11) male monkeys. The machine-learning algorithm random forests classified urine samples as either from normal or T2DM monkeys. The metabolites important for developing the classifier were further examined for their biological significance. Random forests models had a misclassification error of less than 5%. Metabolites were identified based on accurate masses (<10 ppm) and confirmed by tandem mass spectrometry of authentic compounds. Urinary compounds significantly increased (p < 0.05) in the T2DM when compared with the normal group included glycine betaine (9-fold), citric acid (2.8-fold), kynurenic acid (1.8-fold), glucose (68-fold), and pipecolic acid (6.5-fold). When compared with the conventional definition of T2DM, the metabolites were also useful in defining the T2DM condition, and the urinary elevations in glycine betaine and pipecolic acid (as well as proline) indicated defective re-absorption in the kidney proximal tubules by SLC6A20, a Na(+)-dependent transporter. The mRNA levels of SLC6A20 were significantly reduced in the kidneys of monkeys with T2DM. These observations were validated in the db/db mouse model of T2DM. This study provides convincing evidence of the power of metabolomics for identifying functional changes at many levels in the omics pipeline.

Human urinary metabolomic profile of PPARalpha induced fatty acid beta-oxidation

Activation of the peroxisome proliferator-activated receptor alpha (PPARalpha) is associated with increased fatty acid catabolism and is commonly targeted for the treatment of hyperlipidemia. To identify latent, endogenous biomarkers of PPARalpha activation and hence increased fatty acid beta-oxidation, healthy human volunteers were given fenofibrate orally for 2 weeks and their urine was profiled by UPLC-QTOFMS. Biomarkers identified by the machine learning algorithm random forests included significant depletion by day 14 of both pantothenic acid (>5-fold) and acetylcarnitine (>20-fold), observations that are consistent with known targets of PPARalpha including pantothenate kinase and genes encoding proteins involved in the transport and synthesis of acylcarnitines. It was also concluded that serum cholesterol (-12.7%), triglycerides (-25.6%), uric acid (-34.7%), together with urinary propylcarnitine (>10-fold), isobutyrylcarnitine (>2.5-fold), (S)-(+)-2-methylbutyrylcarnitine (5-fold), and isovalerylcarnitine (>5-fold) were all reduced by day 14. Specificity of these biomarkers as indicators of PPARalpha activation was demonstrated using the Ppara-null mouse. Urinary pantothenic acid and acylcarnitines may prove useful indicators of PPARalpha-induced fatty acid beta-oxidation in humans. This study illustrates the utility of a pharmacometabolomic approach to understand drug effects on lipid metabolism in both human populations and in inbred mouse models.

Seedling survival and growth of three ectomycorrhizal caesalpiniaceous tree species in a Central African rain forest

Tree recruitment is determined in part by the survivorship and growth of seedlings. Two seedling cohorts of the three most abundant caesalpiniaceous species forming groves at Korup, Cameroon, were followed from 1995/1997 to 2002, to investigate why Microberlinia bisulcata, the most abundant species, currently has very few recruits compared with Tetraberlinia korupensis and T. bifoliolata. Numbers of seedlings dying, and the heights and leaf numbers of survivors, were recorded on 30 occasions. Survivorship after 2.5 y was 30% for M. bisulcata and 59% for the similar Tetraberlinia spp. together. After 7 y the corresponding values were 4 and 21%. Growth of all species was slow for the first 4 y; but survivors of T. korupensis became 63% taller, as the other species stagnated, by 7 y. The poor recruitment of M. bisulcata was the result of its very low seedling survival. Within species, the tallest seedlings of M. bisulcata and T. bifoliolata, but medium-height ones of T. korupensis, survived longest. This was likely due to higher root allocation in T. korupensis. Seedling dynamics of M. bisulcata and T. korupensis over 7 y accorded well with relative abundances of adult trees; T. bifoliolata is predicted to recruit later.

Herbivores equalize the seedling height growth of three dominant tree species in an African tropical rain forest

Determining the impact of insect herbivores on forest tree seedlings and saplings is difficult without experimentation in the field. Moreover, this impact may be heterogeneous in time and space because of seasonal rainfall and canopy disturbances, or ‘gaps’, which can influence both insect abundance and plant performance. In this study we used fine netting to individually protect seedlings of Microberlinia bisulcata, Tetraberlinia bifoliolata and Tetraberlinia korupensis trees (Fabaceae = Leguminosae) from insects in 41 paired gap-understorey locations across 80 ha of primary rain forest (Korup, Cameroon). For all species, growth in height and leaf numbers was negligible in the understorey, where M. bisulcata had the lowest survival after c. 2 years. In gaps, however, all species responded positively with pronounced above-ground growth across seasons. When exposed to herbivores their seedling height growth was similar, but in the absence of herbivores, M. bisulcata significantly outgrew both Tetraberlinia species and matched their leaf numbers. This result suggests that insect herbivores might play an important role in maintaining species coexistence by mitigating sapling abundance of the more palatable M. bisulcata, which in gaps was eaten the most severely. The higher ratio in static leaf damage of control-to-caged M. bisulcata seedlings in gaps than understorey locations was consistent with the Plant Vigour Hypothesis. This result, however, did not apply to either Tetraberlinia species. For M. bisulcata and T. korupensis, but not T. bifoliolata (the most shade-tolerant species), caging improved relative seedling survival in the understory locations compared to gaps, providing restricted support for the Limiting Resource Model. Approximately 2.25 years after treatments were removed, the caged seedlings were taller and had more leaves than controls in all three species, and the effect remained strongest for M. bisulcata. We conclude that in this community the impact of leaf herbivory on seedling growth in gaps is strong for the dominant M. bisulcata, which coupled to a very low shade-tolerance contributes to limiting its regeneration. However, because gaps are common to most forests, insect herbivores may be having impacts upon functionally similar tree species that are also characterized by low sapling recruitment much more widely than currently appreciated. An implication for the restoration and management of M. bisulcata populations in forests outside of Korup is that physical protection from herbivores of new seedlings where the canopy is opened by gaps, or by harvesting, should substantially increase its subcanopy regeneration, and thus, too, its opportunities for adult recruitment.

Limitation of seedling growth by potassium and magnesium supply for two ectomycorrhizal tree species of a Central African rain forest and its implication for their recruitment

In the ectomycorrhizal caesalpiniaceous groves of southern Korup National Park, the dominant tree species, Microberlinia bisulcata, displays very poor in situ recruitment compared with its codominant, Tetraberlinia bifoliolata. The reported ex situ experiment tested whether availabilities of soil potassium and magnesium play a role. Seedlings of the two species received applications of K and Mg fertilizer in potted native soil in a local shade house, and their responses in terms of growth and nutrient concentrations were recorded over 2 years. Amended soil concentrations were also determined. Microberlinia responded strongly and positively in its growth to Mg, but less to K; Tetraberlinia responded weakly to both. Added Mg led to strongly increased Mg concentration for Microberlinia while added K changed that concentration only slightly; Tetraberlinia strongly increased its concentration of K with added K, but only somewhat its Mg concentration with added Mg. Additions of Mg and K had small but important antagonistic effects. Microberlinia is Mg-demanding and apparently Mg-limited in Korup soil; Tetraberlinia, whilst K-demanding, appeared not to be K-limited (for growth). Added K enhanced plant P concentrations of both species. Extra applied Mg may also be alleviating soil aluminum toxicity, and hence improving growth indirectly and especially to the benefit of Microberlinia. Mg appears to be essential for Microberlinia seedling growth and its low soil availability in grove soils at Korup may be an important contributing factor to its poor recruitment. Microberlinia is highly shade-intolerant and strongly light-responding, whilst Tetraberlinia is more shade-tolerant and moderately light-responding, which affords an interesting contrast with respect to their differing responses to Mg supply. The study revealed novel aspects of functional traits and likely niche-partitioning among ectomycorrhizal caesalps in African rain forests. Identifying the direct and interacting indirect effects of essential elements on tropical tree seedling growth presents a considerable challenge due the complex nexus of causes involved.

Tree architecture in a Bornean lowland rain forest: intraspecific and interspecific patterns

Intraspecific and interspecific architectural patterns were studied for eight tree species of a Bornean rain forest. Trees 5--19 m tall in two 4-ha permanent sample plots in primary forest were selected, and three light descriptors and seven architectural traits for each tree were measured. Two general predictions were made: (1) Slow growing individuals (or short ones) encounter lower light, and have flatter crowns, fewer leaf layers, and thinner stems, than do fast growing individuals (or tall ones). (2) Species with higher shade-tolerance receive less light and have flatter crowns, fewer leaf layers, and thinner stems, than do species with lower shade-tolerance. Shade-tolerance is assumed to decrease with maximum growth rate, mortality rate, and adult stature of a species.