Machine learning algorithms for spatial data. Case studies: Environmental pollution, natural hazards, renewable resources

This paper presents general problems and approaches for the spatial data analysis using machine learning algorithms. Machine learning is a very powerful approach to adaptive data analysis, modelling and visualisation. The key feature of the machine learning algorithms is that they learn from empirical data and can be used in cases when the modelled environmental phenomena are hidden, nonlinear, noisy and highly variable in space and in time. Most of the machines learning algorithms are universal and adaptive modelling tools developed to solve basic problems of learning from data: classification/pattern recognition, regression/mapping and probability density modelling. In the present report some of the widely used machine learning algorithms, namely artificial neural networks (ANN) of different architectures and Support Vector Machines (SVM), are adapted to the problems of the analysis and modelling of geo-spatial data. Machine learning algorithms have an important advantage over traditional models of spatial statistics when problems are considered in a high dimensional geo-feature spaces, when the dimension of space exceeds 5. Such features are usually generated, for example, from digital elevation models, remote sensing images, etc. An important extension of models concerns considering of real space constrains like geomorphology, networks, and other natural structures. Recent developments in semi-supervised learning can improve modelling of environmental phenomena taking into account on geo-manifolds. An important part of the study deals with the analysis of relevant variables and models' inputs. This problem is approached by using different feature selection/feature extraction nonlinear tools. To demonstrate the application of machine learning algorithms several interesting case studies are considered: digital soil mapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis (avalanches, landslides), assessments of renewable resources (wind fields) with SVM and ANN models, etc. The dimensionality of spaces considered varies from 2 to more than 30. Figures 1, 2, 3 demonstrate some results of the studies and their outputs. Finally, the results of environmental mapping are discussed and compared with traditional models of geostatistics.

Managing Climate Insecurity by Ensuring Continuous Capital Accumulation : 'Climate Refugees' and 'Climate Migrants'

Numerous recent reports by non-governmental organisations (NGOs), academics and international organisations have focused on so-called 'climate refugees'. This article examines the turn from a discourse of 'climate refugees', in which organisations perceive migration as a failure of both mitigation and adaptation to climate change, to one of 'climate migration', in which organisations promote migration as a strategy of adaptation. Its focus is the promotion of climate migration management, and it explores the trend of these discourses through two sections. First, it provides an empirical account of the two discourses, emphasising the differentiation between them. It then focuses on the discourse of climate migration, its origins, extent and content, and the associated practices of 'migration management'. The second part argues that the turn to the promotion of 'climate migration' should be understood as a way to manage the insecurity created by climate change. However, international organisations enacts this management within the forms of neoliberal capitalism, including the framework of governance. Therefore, the promotion of 'climate migration' as a strategy of adaptation to climate change is located within the tendencies of neoliberalism and the reconfiguration of southern states' sovereignty through governance.

Assessing the Relationship between the Baseline Value of a Continuous Variable and Subsequent Change over Time

Analyzing the relationship between the baseline value and subsequent change of a continuous variable is a frequent matter of inquiry in cohort studies. These analyses are surprisingly complex, particularly if only two waves of data are available. It is unclear for non-biostatisticians where the complexity of this analysis lies and which statistical method is adequate.With the help of simulated longitudinal data of body mass index in children,we review statistical methods for the analysis of the association between the baseline value and subsequent change, assuming linear growth with time. Key issues in such analyses are mathematical coupling, measurement error, variability of change between individuals, and regression to the mean. Ideally, it is better to rely on multiple repeated measurements at different times and a linear random effects model is a standard approach if more than two waves of data are available. If only two waves of data are available, our simulations show that Blomqvist's method - which consists in adjusting for measurement error variance the estimated regression coefficient of observed change on baseline value - provides accurate estimates. The adequacy of the methods to assess the relationship between the baseline value and subsequent change depends on the number of data waves, the availability of information on measurement error, and the variability of change between individuals.

Calculation of correction factors for ionization chamber measurements with small fields in low-density media.

The quantity of interest for high-energy photon beam therapy recommended by most dosimetric protocols is the absorbed dose to water. Thus, ionization chambers are calibrated in absorbed dose to water, which is the same quantity as what is calculated by most treatment planning systems (TPS). However, when measurements are performed in a low-density medium, the presence of the ionization chamber generates a perturbation at the level of the secondary particle range. Therefore, the measured quantity is close to the absorbed dose to a volume of water equivalent to the chamber volume. This quantity is not equivalent to the dose calculated by a TPS, which is the absorbed dose to an infinitesimally small volume of water. This phenomenon can lead to an overestimation of the absorbed dose measured with an ionization chamber of up to 40% in extreme cases. In this paper, we propose a method to calculate correction factors based on the Monte Carlo simulations. These correction factors are obtained by the ratio of the absorbed dose to water in a low-density medium □D(w,Q,V1)(low) averaged over a scoring volume V₁ for a geometry where V₁ is filled with the low-density medium and the absorbed dose to water □D(w,QV2)(low) averaged over a volume V₂ for a geometry where V₂ is filled with water. In the Monte Carlo simulations, □D(w,QV2)(low) is obtained by replacing the volume of the ionization chamber by an equivalent volume of water, according to the definition of the absorbed dose to water. The method is validated in two different configurations which allowed us to study the behavior of this correction factor as a function of depth in phantom, photon beam energy, phantom density and field size.

Papilledema and obstructive sleep apnea syndrome.

OBJECTIVES: To characterize the pathogenesis and clinical features of optic disc edema associated with obstructive sleep apnea syndrome (SAS). METHODS: A series of 4 patients with SAS and papilledema (PE) underwent complete neuro-ophthalmologic evaluation and lumbar puncture. In 1 patient, continuous 24-hour intracranial pressure (ICP) monitoring was also performed. RESULTS: All 4 patients had bilateral PE that was asymmetric in 2. Three patients had optic nerve dysfunction, asymmetric in 1, unilateral in 2. Daytime cerebrospinal fluid pressure measurements were within normal range. Nocturnal monitoring performed in one patient, however, demonstrated repeated episodes of marked ICP elevation associated with apnea and arterial oxygen desaturation. CONCLUSIONS: We propose that PE in SAS is due to episodic nocturnal hypoxemia and hypercarbia resulting in increased ICP secondary to cerebral vasodilation. In these individuals, intermittent ICP elevation is sufficient to cause persistent disc edema. These patients may be at increased risk for developing visual loss secondary to PE compared with patients with obesity-related pseudotumor cerebri because of associated hypoxemia. The diagnosis of SAS PE may not be appreciated because daytime cerebrospinal fluid pressure measurements are normal and because patients tend to present with visual loss rather than with symptoms of increased ICP.

Continuous monitoring of cerebrovascular pressure reactivity in patients with head injury.

OBJECT: Cerebrovascular pressure reactivity is the ability of cerebral vessels to respond to changes in transmural pressure. A cerebrovascular pressure reactivity index (PRx) can be determined as the moving correlation coefficient between mean intracranial pressure (ICP) and mean arterial blood pressure. METHODS: The authors analyzed a database consisting of 398 patients with head injuries who underwent continuous monitoring of cerebrovascular pressure reactivity. In 298 patients, the PRx was compared with a transcranial Doppler ultrasonography assessment of cerebrovascular autoregulation (the mean index [Mx]), in 17 patients with the PET-assessed static rate of autoregulation, and in 22 patients with the cerebral metabolic rate for O(2). Patient outcome was assessed 6 months after injury. RESULTS: There was a positive and significant association between the PRx and Mx (R(2) = 0.36, p < 0.001) and with the static rate of autoregulation (R(2) = 0.31, p = 0.02). A PRx > 0.35 was associated with a high mortality rate (> 50%). The PRx showed significant deterioration in refractory intracranial hypertension, was correlated with outcome, and was able to differentiate patients with good outcome, moderate disability, severe disability, and death. The graph of PRx compared with cerebral perfusion pressure (CPP) indicated a U-shaped curve, suggesting that too low and too high CPP was associated with a disturbance in pressure reactivity. Such an optimal CPP was confirmed in individual cases and a greater difference between current and optimal CPP was associated with worse outcome (for patients who, on average, were treated below optimal CPP [R(2) = 0.53, p < 0.001] and for patients whose mean CPP was above optimal CPP [R(2) = -0.40, p < 0.05]). Following decompressive craniectomy, pressure reactivity initially worsened (median -0.03 [interquartile range -0.13 to 0.06] to 0.14 [interquartile range 0.12-0.22]; p < 0.01) and improved in the later postoperative course. After therapeutic hypothermia, in 17 (70.8%) of 24 patients in whom rewarming exceeded the brain temperature threshold of 37 degrees C, ICP remained stable, but the average PRx increased to 0.32 (p < 0.0001), indicating significant derangement in cerebrovascular reactivity. CONCLUSIONS: The PRx is a secondary index derived from changes in ICP and arterial blood pressure and can be used as a surrogate marker of cerebrovascular impairment. In view of an autoregulation-guided CPP therapy, a continuous determination of a PRx is feasible, but its value has to be evaluated in a prospective controlled trial.

Can accelerometry accurately predict the energy cost of uphill/downhill walking?

To evaluate whether an activity monitor based on body acceleration measurement can accurately assess the energy cost of the human locomotion, 12 subjects walked a combination of three different speeds (preferred speed +/- 1 km/h) and seven slopes (-15 to +15% by steps of 5%) on a treadmill. Body accelerations were recorded using a triaxial accelerometer attached to the low back. The mean of the integral of the vector magnitude (norm) of the accelerations (mIAN) was calculated. VO2 was measured using continuous indirect calorimetry. When the results were separately analysed for each incline, mIAN was correlated to VO2 (average r = 0.87, p<0.001, n = 36). VO2 was not significantly correlated to mIAN when data were globally analysed (n = 252). Large relative errors occurred when predicted VO2 (estimated from data of level walking) was compared with measured VO2 for different inclines (-53% at +15% incline, to +55% at -15% incline). It is concluded that without an external measurement of the slope, the standard method of analysis of body accelerations cannot accurately predict the energy cost of uphill or downhill walking.

Potential evidence of parasite avoidance in an avian malarial vector

Epidemiological studies of malaria or other vector-transmitted diseases often consider vectors as passive actors in the complex life cycle of the parasites, assuming that vector populations are homogeneous and vertebrate hosts are equally susceptible to being infected during their lifetime. However, some studies based on both human and rodent malaria systems found that mosquito vectors preferentially selected infected vertebrate hosts. This subject has been scarcely investigated in avian malaria models and even less in wild animals using natural host-parasite associations. We investigated whether the malaria infection status of wild great tits, Parus major, played a role in host selection by the mosquito vector Culex pipiens. Pairs of infected and uninfected birds were tested in a dual-choice olfactometer to assess their attractiveness to the mosquitoes. Plasmodium-infected birds attracted significantly fewer mosquitoes than the uninfected ones, which suggest that avian malaria parasites alter hosts' odours involved in vector orientation. Reaction time of the mosquitoes, that is, the time taken to select a host, and activation of mosquitoes, defined as the proportion of individuals flying towards one of the hosts, were not affected by the bird's infection status. The importance of these behavioural responses for the vector is discussed in light of recent advances in related or similar model systems.

Single-injection or continuous femoral nerve block for total knee arthroplasty?

BACKGROUND: The ideal local anesthetic regime for femoral nerve block that balances analgesia with mobility after total knee arthroplasty (TKA) remains undefined. QUESTIONS/PURPOSES: We compared two volumes and concentrations of a fixed dose of ropivacaine for continuous femoral nerve block after TKA to a single injection femoral nerve block with ropivacaine to determine (1) time to discharge readiness; (2) early pain scores and analgesic consumption; and (3) functional outcomes, including range of motion and WOMAC scores at the time of recovery. METHODS: Ninety-nine patients were allocated to one of three continuous femoral nerve block groups for this randomized, placebo-controlled, double-blind trial: a high concentration group (ropivacaine 0.2% infusion), a low concentration group (ropivacaine 0.1% infusion), or a placebo infusion group (saline 0.9% infusion). Infusions were discontinued on postoperative Day (POD) 2. The primary outcome was time to discharge readiness. Secondary outcomes included opioid consumption, pain, and functional outcomes. Ninety-three patients completed the study protocol; the study was halted early because of unanticipated changes to pain protocols at the host institution, by which time only 61% of the required number of patients had been enrolled. RESULTS: With the numbers available, the mean time to discharge readiness was not different between groups (high concentration group, 62 hours [95% confidence interval [CI], 51-72 hours]; low concentration group, 73 hours [95% CI, 63-83 hours]; placebo infusion group 65 hours [95% CI, 56-75 hours]; p = 0.27). Patients in the low concentration group consumed significantly less morphine during the period of infusion (POD 1, high concentration group, 56 mg [95% CI, 42-70 mg]; low concentration group, 35 mg [95% CI, 27-43 mg]; placebo infusion group, 48 mg [95% CI, 38-59 mg], p = 0.02; POD 2, high concentration group, 50 mg [95% CI, 41-60 mg]; low concentration group, 33 mg [95% CI, 24-42 mg]; placebo infusion group, 39 mg [95% CI, 30-48 mg], p = 0.04); however, there were no important differences in pain scores or opioid-related side effects with the numbers available. Likewise, there were no important differences in functional outcomes between groups. CONCLUSIONS: Based on this study, which was terminated prematurely before the desired sample size could be achieved, we were unable to demonstrate that varying the concentration and volume of a fixed-dose ropivacaine infusion for continuous femoral nerve block influences time to discharge readiness when compared with a conventional single-injection femoral nerve block after TKA. A low concentration of ropivacaine infusion can reduce postoperative opioid consumption but without any important differences in pain scores, side effects, or functional outcomes. These pilot data may be used to inform the statistical power of future randomized trials. LEVEL OF EVIDENCE: Level II, therapeutic study. See Guidelines for Authors for a complete description of levels of evidence.

Human endometrial carcinomas serially transplanted in nude mice and established in continuous cell lines.

Three out of five human endometrial carcinomas were successfully grafted into nude mice (BALB/c/nu/nu). Two of these tumors could be maintained by serial transplantation. The morphological characteristics displayed by the grafted tumors were comparable to those of the original carcinomas. Permanent cell lines were established from these two tumors. Reinjection of cells grown in vitro into nude mice produced nodules of identical histology as compared to original solid transplants. The influence of medroxyprogesterone acetate on tumor growth in vivo and cell proliferation in vitro was studied. This hormonal treatment did not produce any significant effect on tumor cells, either in vitro or in vivo, for the two endometrial carcinomas. After medroxyprogesterone administration, a slight but non-significant growth inhibition of the tumor cells in vitro was observed and the tumor transplants in vivo did not appear to be influenced. The experiments illustrate the possible use of this model for testing potential anti-cancer agents.

Lentiviral Vector Tropism In Degenerating Retinas

Introduction: In normal mice, lentiviral vector (LV) shows a great efficiency to infect the RPE cells, but transduces retinal neurons more efficiently during development. Here, we investigated the tropism of LV in the degenerating retina of mice, knowing that the retina structure changes during degeneration. We postulated that the viral transduction would be increased by the alteration of the interphotoreceptor matrix (IPM). We tested two different LV-pseudotypes using the VSVG and the Mokola envelopes. Methods: Subretinal injections were performed in wild-type (C57/Bl6) and rhodopsin knockout (Rho-/-) mice. We injected LV-VSVG-EFS-GFPII into 3.3-4.9 month old mice and LV-VSVG-Rho-GFP into 1-1.4 month old mice to target the photoreceptors (PR). LV-MOK-CMV-GFP was injected into 2.4-3.3 months old mice. We sacrificed the animals one week post injection, used immunohistochemistry to identify the transduced cells, and investigated the OLM integrity. Results: Using LV-VSVG-EFS-GFPII into 3.3-4.9 months mice, we observed significant retinal and RPE transduction in Rho-/- mice. However, the retinas showed transduction mainly at the injection's site. We mostly observed GFP+ cells having a Müller cell morphology. Using LV-MOK-CMV-GFP into 2.4-3.3 months mice, we evidenced the same pattern of viral infection, but with more Müller cells targeted by the virus. Using LV-VSVG-Rho-GFP into 1-1.4 month old mice, we don't note any difference between Rho-/- and wild-type mice for transduced cells. The IPM stained with ZO1 appears irregular into the 4.9 months old Rho-/- mice; for the youngest mice (Rho-/- and C57/Bl6), there is no modification of the IPM. Conclusion: The degeneration improves retinal cells transduction due to the alteration of the IPM in old Rho-/- mice. Müller cells seem (by morphological evidences) to be the principal cells expressing the transgene. The LV with Mokola envelope can transduce Müller cells in a degenerating retina with an intact IPM. In 1 month old mice, the degeneration doesn't enhance the transduction in rod PR probably because the IPM is not yet altered. The possibility to target photoreceptors at a later stage of the degeneration is under investigation.

Bilateral sequential nonarteritic anterior ischemic optic neuropathy: a comparison of visual outcomes in fellow eyes using quantitative analysis of goldmann visual fields.

OBJECTIVE To better define the concordance of visual loss in patients with nonarteritic anterior ischemic optic neuropathy (NAION). METHODS The medical records of 86 patients with bilateral sequential NAION were reviewed retrospectively, and visual function was assessed using visual acuity, Goldmann visual fields, color vision, and relative afferent papillary defect. A quantitative total visual field score and score per quadrant were analyzed for each eye using the numerical Goldmann visual field scoring method. RESULTS Outcome measures were visual acuity, visual field, color vision, and relative afferent papillary defect. A statistically significant correlation was found between fellow eyes for multiple parameters, including logMAR visual acuity (P = .01), global visual field (P < .001), superior visual field (P < .001), and inferior visual field (P < .001). The mean deviation of total (P < .001) and pattern (P < .001) deviation analyses was significantly less between fellow eyes than between first and second eyes of different patients. CONCLUSIONS Visual function between fellow eyes showed a fair to moderate correlation that was statistically significant. The pattern of vision loss was also more similar in fellow eyes than between eyes of different patients. These results may help allow better prediction of visual outcome for the second eye in patients with NAION.