Innovative Projects Realized

Statistical challenges in individual patient data meta analysis

In a typical meta analysis, estimates of the parameter of interest (e.g. odds ratio) are extracted from the literature or by contacting researchers and pooled together. In contrast, for IPD-MA line-by-line patient data are obtained from each study.
IPD-MA data permit researchers to define exposures and outcomes consistently across studies, and to analyze the association of interest consistently (e.g. adjusting for the same confounders), which may minimize heterogeneity. IPD-MA are of particular value for synthesis of observational studies. In traditional MAs based on published data, it is difficult to investigate differences between study results, to adjust for differences in populations across studies and to pool effects that have been adjusted for other variables.
IPD-MA have higher power than meta-regression to detect covariate-treatment interactions, and are preferable when the aim is to estimate interactions with patient-level covariates. IPD-MA are not prone to ecological bias, because patient-level data are not aggregated.
While in some cases IPD-MA and aggregate data MA may give similar results, this is unlikely when evaluating treatment-covariate interactions, incorporating nonlinear relationships, when trials are small, and there is heterogeneity across trials, and particularly for pooling of non-randomized studies that may need to adjust for several confounders.
For these reasons, IPD-MA are considered the gold-standard of MA, despite the complexity and cost of collecting the data, and are published with increasing frequency.

For IPD-MA, two broad analytic strategies (one- and two-step approaches) are possible; both preserve the clustering of subjects within studies, comparability of study arms, and may be either fixed or random. A fixed effects analysis assumes that the estimated effect is the same across all studies; a random effects analysis assumes that the estimated effect varies across studies due to differences in patient populations, study procedures, etc.
A one-step approach offers more flexibility to explore the differences that may exist between patients in the same study as well as across studies. Overall, most statisticians and meta-analyst methodologists agree that a one-step approach is better and more flexible than a two step approach.

While the biggest drawback of an IPD-MA is the time/expense to assemble the data, our experience demonstrates that despite many advantages, the wide range of methods used and lack of a standardized data analytics plan is also a serious problem. Compared to conventional MA, methods for IPD-MA are described as “more complex and not well-known”,14 maybe due to several open questions about the analytic plan. Next, we address these.

IPD-MA are the gold standard of MA, offering many opportunities for the sophisticated modelling of dose-response curves, effect of and adjustment for patient-level covariates as well as treatment interactions when a one-step approach is used. However, several methodologic challenges remain. This project aims to investigate these methodologic challenges and propose the best data analytic strategies. This project will introduce students to research in Biostatistics – that is at the intersection of Statistics and Epidemiology.

Hydrogen trapping at nanoscale alloy carbides in high-strength steels

Hydrogen, when dissolved in metal or alloy lattice, can render the otherwise ductile material brittle. This phenomenon, namely hydrogen embrittlement (HE), has been a persistent obstacle for the application and development of high strength steels. One effective means to enhance the high strength steel’s resistance to HE is to introduce nanoscale alloy carbides (NACs) into the microstructure, evidenced by the increased emission temperature of hydrogen and delayed fracture under the same hydrogen charging condition. The role of NACs in HE is often attributed to them trapping hydrogen. Nonetheless, with NACs being nanoscale in size, their trapping effects on hydrogen are difficult to quantify in experiments. In this project, we employ first-principle calculations which provide an accurate way to examine NACs and their interplay with hydrogen with atomic resolution. Several candidate NACs in high-strength steels will be considered. For each of them, the hydrogen adsorption energetics and diffusion kinetics at the NAC will be investigated. The individual roles of the NAC induced coherent lattice strain and NAC/metal interface structure will be clarified. In addition, preliminary studies will be conducted on incoherent NAC/metal interfaces. The results obtained will provide quantitative assessments of the role of NACs on hydrogen trapping, offering important bottom-up insights for the microstructure engineering of high-strength steels against HE.

Magnetometry with spin-based sensors

The summer intern will participate in a project aimed at using single defect centers in diamond to map out the magnetic fields generated by driven ferromagnetic circuits. Nanoscale magnetic circuits are critically important to computer hardware, especially in applications that use ferromagnetic domains to encode information. For example, magnetic random access memory (MRAM) could potentially replace standard DRAM while drastically reducing the demands for power. Understanding the dynamics of driven ferromagnetic circuits thus has important applications both in fundamental science and in industry. One challenge is to understand the role of spatial inhomogeneities, and one would thus seek to map out the magnetization of the device on the nanoscale – and ideally also watch its time evolution. However, there are few techniques that can measure magnetic fields with good sensitivity and excellent spatial and temporal (or, alternately, spectral) resolution.

One possibility is to fabricate the ferromagnetic circuit on the surface of a substrate embedded with tiny magnetic sensors. In particular, the nitrogen-vacancy (NV) defect center in diamond can be used to sense magnetic fields because its spin (which responds to magnetic fields) can be detected via the fluorescence intensity of the defect. Each defect can thereby reveal the magnetic field at its location, with the potential for nanoscale spatial resolution. Moreover, by manipulating the defect, one can select the time at which it senses the magnetic field or (more commonly) the frequency of magnetic fields that affect it. The central goal of this project is thus to use a diamond substrate embedded with NV centers to map out the driven dynamics of a ferromagnetic circuit fabricated on the diamond surface.

This project is in its initial stages, and the first goal for the project will be to observe resonant features in the circuit dynamics by correlating transport measurements (e.g. the resistance of the circuit as a function of the frequency of the AC current that drives it) with measurements of NV defects. Currently, we have begun implantation of defects into diamond substrates, and will begin fabrication of simple magnetic circuits this fall. Next summer, the first devices will likely be functional. The intern will help to push this project forward by developing a specific aspect of the apparatus or data analysis.

Development of Biocompatible and bidegradable Mg alloy stents for cardiovascular implant applications

Cardiovascular diseases are the leading cause of premature death in developed countries like Canada. Many such diseases are now treated by inserting medical devices (e.g. intracardiac occlusive devices, intra-vascular stents) implanted through percutaneous or minimally invasive approach. Most implants are of permanent materials (stainless steel, cobalt) that may cause long-term erosion, irritation, inflammation, perforation, infection and lack of growth. Also, the surgical extraction or reintervention in the presence of these devices is difficult or impossible. Bio-degradable implants (e.g. stents) would eliminate these problems correcting the underlying pathology and ”fading away” once the problem is fixed. The new Mg alloy implants (stents) targeted in this research are expected to be significantly more biocompatible and function-efficient than those currently available. This will minimize post-implantation complications & the need for re-intervention, improve the quality of a patient’s life, & decrease health care costs. In this multidisciplinary research, we integrated expertise in materials engineering, electrochemistry, cell biology and cardio-vascular clinical science & practice to develop optimal magnesium (Mg) alloy (s) for intravascular implant fabrication. The optimal alloys are developed in a step-wise fashion to attain bio-corrosion resistance, bio-compatibility and mechanical integrity (strength, ductility, bendability). In this part of the research the alloys are tested in simulated-body fluid to evaluate bio-corrosion by monitoring mass loss, and the pH of the solution. Mechanical testing is conducted via tensile and three-point bending tests.

The impact of climate change on maternal-child health in Canada

Climate change is a fairly new phenomenon and its impact on health is poorly understood. Extreme weather events are becoming increasingly common, but the effect on vulnerable populations such as women and children is largely unknown. One particular problem that limits research on the impacts of extreme weather on health is the lack of indicators to help define extreme weather. Although research shows that heat waves defined as temperatures over 32 degrees Celsius for three consecutive days in Quebec can shorten the length of pregnancy, indicators to capture other types of climate stresses are lacking, including indicators for extreme precipitation or winter storms. Data on past weather patterns are available from Environment Canada, but they have yet to be used in research on perinatal health impacts of climate change in Quebec. This project aims to better understand the impact of climate change on maternal-child health by developing indicators (ie variables) to describe extreme weather in Quebec, Canada. This includes developing indicators to measure intense heat waves, precipitation, and severe winter storms that can be used for research on maternal-child health.

Evaluation of a robotic orthosis for the rehabilitation of the locomotion in children suffering from cerebral palsy.

This project is about the use of a robotic device for the rehabilitation of locomotion in children suffering from cerebral palsy. This multidisciplinary research will grant the candidate an experience in biomechanical research, with the use of technologies at the edge of clinical rehabilitation.

Cerebral palsy is the main factor of locomotion troubles in children. In North America, about 3 children in 1000 are born with this disorder . Current rehabilitation methods consist in the repetition of motor tasks such as walking, in order to reconstruct the neural network. Movements are often assisted by physiotherapists. Nevertheless, the use of robotic devices is emerging to assist the physiotherapist in locomotion rehabilitation.

Some studies highlighted that robotic orthoses reproducing the walking pattern on a treadmill are efficient to improve the locomotion in people suffering from cerebral palsy. Our research center in pediatric rehabilitation has recently acquired a Lokomat: a gait therapy device that enables to reproduce the walk on a treadmill with a robotic gait orthosis which can relieve bodyweight. Although the effect of Lokomat training in adult patients is documented, little investigation has been conducted on a children population. The benefits in paediatric rehabilitation remain unknown.

Some studies pointed out that the hip kinematics and the level of thigh muscle activation was different during or after training with Lokomat in healthy and cerebral palsied children. However, due to the small number of participants and the reliability of the kinematic measurements, results should be regarded with caution. Moreover, to the best of our knowledge, the effect of Lokomat training on the contribution of each lower limb joint during walking in cerebral palsied children has never been studied. Finally, even if the Lokomat enables a relief of bodyweight from 0% to 100%, there are no actual recommendations concerning this adjustment. However, rehabilitation could be optimized through a subject-specific training protocol.

Consequently, it seems relevant to get a better understanding of the factors underlying the locomotion of cerebral palsied children in order to optimize the Lokomat training and the rehabilitation efficiency.
The candidate will work on two projects led by two postdoctoral fellows. The objectives are to:
1) Evaluating the ability of the Lokomat to reproduce the kinematic of walking performed by a healthy child;
2) Determining the effects of Lokomat training on joint kinematics and kinetics, as well as the motor control mechanisms (through electromyography and electroencephalography) underlying the relearning of locomotion in cerebral palsied children

Recommendations deriving from this study will directly benefit the rehabilitation of cerebral palsy in children. Moreover, despite the clinical relevancy of the Lokomat, this apparatus is unusual in Canada, only one team located at the University of Vancouver (BC), has already published a scientific article using the Lokomat. Hence, this project will also give the opportunity to the candidate to get involved in a multidisciplinary team and to be trained on a new emerging rehabilitation device.

Musculoskeletal modeling for estimating muscle force in human: C++ optimization strategies for real-time analysis

Thanks to recent research in biomechanics, the estimation of human muscle forces during a dynamic movement becomes more and more feasible. Applications of such tools can be directed at improving sport performance, diagnoses of musculoskeletal diseases and monitoring progress made in rehabilitation. Still, due to the human body over-actuation (more than one muscle can produce the same action), a scientific challenge remains.

Indeed, the estimation of muscle forces relies on an optimization processes such as static optimization or optimal control. Contrary to static optimization which is based on inverse flow and is sensitive to kinematic noise, optimal control follows a direct flow. The main advantage of the forward procedure is that both muscle activation and kinematics states take part in the optimization process while the static optimization (inverse flow) relies on the hypothesis that kinematics is noiseless and perfectly determined. In reality, the latter is rarely, if not never fulfilled. On the other hand, the number of optimized parameters in an optimal control procedure drastically increases with the number of degrees of freedom of the kinematics model, the number of muscles of the muscular model and the number of time nodes to be optimized. Thus, an optimization process may run some functions over a billion times for the simulation of a simple movement. In this condition, each and every millisecond lost in useless calculation time can lead to minutes and hours waiting in front of the computer.

The goal of this project is to monitor in detail the number of calls and the computational time taken by each function and subfunction involved in the process of an optimal control simulation. As an overview of the flow for one call of the main function, controls (i.e. muscular excitation; inputs of the function) are converted to muscular forces. Then, all the muscular forces are combined to compute joint torques using contact Jacobian matrix. From these joint torques, joint accelerations (outputs), using a forward dynamics function provided by the Rigid Body Dynamics Library (RBDL), are computed. It is to be noted that all matrix calculations are done using the Eigen library.

When the full monitoring is done and validated, the second goal of this project will be to improve and optimize the slowest and/or to reduce the number of calls for the most called functions in order to minimize the global computing time. As an example, the mass matrix inversion while calculating the generalized joint acceleration may be improved by implementing or developing sparse inversion matrix algorithms since this matrix is largely sparse.

This project is part of a larger project about optimal movement synthesis, where the goal is to reproduce from scratch, improve and propose new personalized techniques in sports movement such as diving and jumping for high level athletes.

Role of metabolic inflammation in depressive behavior

Obesity and diabetes are significant risk factors for mood disorders. Diabetes doubles the incidence of depression and as many as 1 in 3 people with diabetes has depressed mood at a level that diminishes functioning, glycemic control, adherence to treatment and increases the risk of diabetes complications. Over consumption of energy-dense foods significantly contribute to the development of obesity and type 2 diabetes (T2D). Palatable high-fat foods are rewarding and can provide short-term relief from stress and negative emotional states. On the other hand, we have shown that prolonged intake of a high-fat diet (HFD) leading to obesity in mice promotes depressive-like behavior to implicate excessive caloric intake as an important element tying obesity to depression. The magnitude of depressive-like behavior was positively associated with neuroplastic adaptations including increased brain-derived neurotrophic factor (BDNF) and phospho-CREB in the nucleus accumbens (NAc), a limbic brain region strongly tied to hedonic and motivational deficits found in depression. BDNF and CREB upregulation in the NAc and are implicated in the etiology of depressed mood and are modulated by nuclear factor kappa B (NF?B), a transcription factor that plays a key role in the in the cellular response to stress, cytokine production and neuroplasticity. Our new data show that prolonged intake of a saturated HFD, but not an isocaloric monounsaturated HFD, leading to obesity, hyperglycemia and glucose intolerance elicits depressive-like behavior and neuroinflammatory responses in the NAc that include activation of NF?B.
Several lines of evidence implicate immune responses in the pathophysiology of depression . Situations favorable to prolonged systemic inflammation such as obesity and T2D provide an avenue thorough which neuroinflammatory responses can be propagated and sustained. Indeed, inflammatory responses are evident in hypothalamic nuclei in high-fat fed mice and they contribute to obesity and glucose intolerance in a manner that depends on IKK?/NF?B signaling. Our new results demonstrate that the saturated HFD regimen, besides promoting peripheral metabolic inflammation, triggers neuroinflammation in the NAc including reactive gliosis (brain immune cell “microglia” activation and astrogliosis), increased expression of IL1? and TNF?, activation of NF?B and leukocyte infiltration. Together, our new findings clearly demonstrate that a saturated HFD leading to obesity, hyperinsulinemia and glucose intolerance elicits metabolic neuroinflammation in brain nuclei strongly implicated in the control of mood and motivation. Our global objective is to identify the neural processes by which high-fat feeding, weight gain and T2D lead to depressive-like behavior. We hypothesize that saturated high-fat feeding and the development of hyperglycemia progressively leads to metabolic neuroinflammation in the NAc to elicit cell injury, neuroplastic changes (increased BDNF and CREB) and depressive behavior in an IKK?/NF?B-dependent manner. The proposed project will test two specific hypotheses:

Exploitation of second-order information in models estimation

With the advances in big data analyses, the estimation of models is a more and more important issue. Two main estimation techniques are prevalent: least-squares minimization and likelihood maximization. These problems are typically solved using iterative optimization techniques, but given a model and an initial, the time required to obtain satisfactory estimates can be prohibitive due to the large number of observations, especially when non-linearities or/and heterogeneity have to be taken into account. Various strategies have been proposed to speed up the estimation time, especially when the models are twice-continuously differentiable. The first possibility is to increase the number of observations considered in the optimization only when close to the solution, as the first iterations solutions will be discarded. The question remains how to increase the sample during the estimation process, and how to select the observations to add. The second possibility is to exploit the mathematical structure of the problems, especially when using optimization techniques relying on quadratic models. In both cases, if the model is properly identified, the Hessian of the function to optimization can be related to the outer product of the scores, that is the individual gradient contributions to the average gradient of the objective function. However, model misspecification can lead to non-convergent candidate solutions sequences, as the constructed quadratic models do not properly approximate the objective function when close to the true parameters values to estimate. In order to circumvent this limitation, Hessian correction techniques, based on the secant equation, have been proposed. This has lead to Gauss-Newton algorithm in case of the non-linear least-squares estimation problem, and similar techniques for maximum likelihood. In the latter cases, we have however observed that for complex models, corrections based on positive-definite Hessian approximations, as BFGS, can perform quite poorly, while other techniques, e.g. SR1, can perform much better. Our current intuition is that since the outer product of the scores is already positive definite, to constraint the eigenvalues of the correction matrix to be positive can prevent to have an adequate correction. We also expect that a similar effect is prevalent for least-square estimation. The goal of the project is to validate this explanation and provide more adequate guidance in Hessian correction when estimating models by least-squares minimization or maximum likelihood estimation using standard non-linear optimization algorithms, that is line-search methods or trust-region algorithms. The applications to variants, as retrospective techniques, will also be considered. The impact on some adaptive sampling strategies, that allow to vary the number of observations used in the objective function at each iteration, will also be considered.

Sustainable development: Environment, Economy, Human Rights (English or Français) (New)

A synthesis of current social, cultural and legal issues associated with international sustainable development law. How these aspects interconnect to impede or support sustainable development especially in particularly vulnerable populations such as indigenous peoples.

Simulation of electrocardiograms in a computer model of the heart and torso

Atrial fibrillation is the most frequent rhythm disorder in humans (nearly 250,000 patients in Canada). It often leads to severe complications such as heart failure and stroke. Diagnosis of this arrhythmia is mainly performed through the inspection of electrical signal recordings (electrograms and electrocardiograms). To develop and validate new diagnostic tools, it is necessary to understand the link between what the cardiologist observes (these electrical signals) and what is going wrong in the heart (the underlying cardiac pathology).

In parallel with the dramatic increase in computer power over the last few decades, computer models of cardiac electrical activity have evolved from small strings of cells to a detailed description of the whole heart. Integrating information from the molecular scale to the whole organ, our models can not only simulate arrhythmias and investigate mechanisms but also can evaluate diagnostic and therapeutic approaches. Used in combination with experimental and clinical research, computer modeling is expected to play an increasing role in the interpretation of biomedical measurements.

We create three-dimensional virtual models of the human atria based on anatomical, histological and electrophysiological data. In these models, conditions are set up that trigger and maintain an arrhythmia, as inspired by clinical observations and physiological hypotheses. A variety of conditions are simulated to reproduce different diseased states of increasing severity. The evolution of the electrical activity generated by the heart during an arrhythmia is simulated. Then, electrical signals obtained from computer simulations, animal experiments and patients can be analyzed and compared.

Antibiofilm molecules active agains staphylococci / Molécules antibiofilm actives contre les staphylocoques (Nouveau)

Bacteria within biofilms can withstand the host immune responses, and they are significantly more tolerant to antibiotics and disinfectants. Our laboratory found that some coagulase-negative staphyloccal (CNS) isolates can efficiently block biofilm formation by other CNS species or Staphylococcus aureus. Hypothesis: Some CNS isolates produce antibiofilm molecules that can lead to the development of new drugs and/or new strategies to control staphylococcal infection through biofilm dispersion or prevention. Objectives : (i) to determine the mechanism underlying the antibiofilm activity of CNS and (ii) the spectrum of activity on different bacterial species from our collection. Methodology : Biofilm production by bacterial isolates will be measured after growth in microtiter plates and quantification of the biofilm layer by using a standard staining procedure. We will investigate the ability of CNS to secrete enzymes with protease, lipase and/or nuclease activities using standard microbiological and biochemical methods. We will investigate if CNS can produce antibacterial compounds, known as bacteriocins. The spectrum of antibiofilm activity will be determined using a collection of CNS and S. aureus isolates, and of other bovine mastitis pathogens such as Steptococcus uberis, Strep. dysgalactiae, and E. coli. These antibiofilm molecules, used alone or in combination with an antibacterial agent, may represent a novel strategy to control bacterial infections.

————————————–

Les biofilms bactériens représentent un problème important en santé animale et en santé humaine à cause de leur résistance accrue aux antibiotiques et aux désinfectants. Notre laboratoire a mis en évidence que certaines souches de staphylocoque à coagulase négative (SCN) pouvaient bloquer d’une manière très efficace la formation de biofilms par d’autres staphylocoques dont S. aureus. Nous émettons l’hypothèse que ces souches de SCN produisent des molécules antibiofilm ce qui pourrait mener au développement de nouvelles drogues ou de nouvelles stratégies pour contrôler la formation de biofilm. Nos objectifs sont: 1)