Marcelo Lobosco

An evaluation of cJava system architecture

We propose a new distributed run-time environment, which we called cJava, that enables multithread Java applications to execute in clusters transparently. Our implementation of cJava supports the distributed shared memory (DSM) which required significant extensions to the original Java virtual machine (JVM). First, a distributed object manager was incorporated to the JVMs memory management subsystem for creating a global object space. Second, synchronized accesses to the global object space were extended so that they could use the lock() and unlock() primitives that cJavas DSM supports. Third, cJava adapted the thread subsystem to enable remote creation and global monitors. Last, a subsystem/or remote signaling was added to the original JVM. The main advantage of cJava is that it can execute existing multithread Java applications straightaway. Most importantly, our results of cJavas performance across several benchmarks show that cJava offers an efficient run-time system for executing transparently multithread Java applications in clusters.

On the computational modeling of the innate immune system

In recent years, there has been an increasing interest in the mathematical and computational modeling of the human immune system (HIS). Computational models of HIS dynamics may contribute to a better understanding of the relationship between complex phenomena and immune response; in addition, computational models will support the development of new drugs and therapies for different diseases. However, modeling the HIS is an extremely difficult task that demands a huge amount of work to be performed by multidisciplinary teams. In this study, our objective is to model the spatio-temporal dynamics of representative cells and molecules of the HIS during an immune response after the injection of lipopolysaccharide (LPS) into a section of tissue. LPS constitutes the cellular wall of Gram-negative bacteria, and it is a highly immunogenic molecule, which means that it has a remarkable capacity to elicit strong immune responses. We present a descriptive, mechanistic and deterministic model that is based on partial differential equations (PDE). Therefore, this model enables the understanding of how the different complex phenomena interact with structures and elements during an immune response. In addition, the models parameters reflect physiological features of the system, which makes the model appropriate for general use.

A three-dimensional computational model of the innate immune system

The Human Immune System is a complex system responsible for protecting the organism against diseases. Although understanding how it works is essential to develop better treatments against diseases, its complexity makes this task extremely hard. In this work a three-dimensional mathematical and computational model of part of this system, the innate immune system, is presented. The high computational costs associated to simulations lead the development of a parallel version of the code, which has achieved a speedup of about 72 times over its sequential counterpart.

Automatic Segmentation of Cardiac MRI Using Snakes and Genetic Algorithms

In this work we study and implement techniques for the automatic segmentation of cardiac Magnetic Resonance Images. The methods are based on the active contours algorithm called Snakes, which are adapted and tailored to the specific task of automatic segmentation of the left ventricle of the heart in Magnetic Resonance Images. We propose a new external force to improve the convergence of the Snakes method. In addition, a genetic algorithm is used to find the best set of configuration parameters for the Snakes method. The algorithms are implemented in Java and threads are used to explore data parallelism on shared-memory machines. Tests are performed on 150 short-axis images acquired from two healthy volunteers. Preliminary results suggest the proposed methods are promising and with further development and validation may be used, for instance, for the automatic calculation of cardiac ejection fraction.

Simulation of Ectopic Pacemakers in the Heart: Multiple Ectopic Beats Generated by Reentry inside Fibrotic Regions

The inclusion of nonconducting media, mimicking cardiac fibrosis, in two models of cardiac tissue produces the formation of ectopic beats. The fraction of nonconducting media in comparison with the fraction of healthy myocytes and the topological distribution of cells determines the probability of ectopic beat generation. First, a detailed subcellular microscopic model that accounts for the microstructure of the cardiac tissue is constructed and employed for the numerical simulation of action potential propagation. Next, an equivalent discrete model is implemented, which permits a faster integration of the equations. This discrete model is a simplified version of the microscopic model that maintains the distribution of connections between cells. Both models produce similar results when describing action potential propagation in homogeneous tissue; however, they slightly differ in the generation of ectopic beats in heterogeneous tissue. Nevertheless, both models present the generation of reentry inside fibrotic tissues. This kind of reentry restricted to microfibrosis regions can result in the formation of ectopic pacemakers, that is, regions that will generate a series of ectopic stimulus at a fast pacing rate. In turn, such activity has been related to trigger fibrillation in the atria and in the ventricles in clinical and animal studies.

A GPU-based heart simulator with mass-spring systems and cellular automaton

This work proposes an electro-mechanical simulator of the cardiac tissue, so that its main feature is the low computational cost. This is necessary to run real-time simulations and on the fly applications. In order to achieve this, we used cellular automata and mass-spring systems to model the cardiac behavior, and furthermore parallelize the code to run in graphics processing unit (GPU) with compute unified device architecture. Sequentially, our simulator was quite faster than traditional partial differential equations simulators. In addition, we performed different load tests to evaluate our code behavior in GPUs, and spotted its potentials and bottlenecks.

Simulations of Complex and Microscopic Models of Cardiac Electrophysiology Powered by Multi-GPU Platforms

Key aspects of cardiac electrophysiology, such as slow conduction, conduction block, and saltatory effects have been the research topic of many studies since they are strongly related to cardiac arrhythmia, reentry, fibrillation, or defibrillation. However, to reproduce these phenomena the numerical models need to use subcellular discretization for the solution of the PDEs and nonuniform, heterogeneous tissue electric conductivity. Due to the high computational costs of simulations that reproduce the fine microstructure of cardiac tissue, previous studies have considered tissue experiments of small or moderate sizes and used simple cardiac cell models. In this paper, we develop a cardiac electrophysiology model that captures the microstructure of cardiac tissue by using a very fine spatial discretization (8 μm) and uses a very modern and complex cell model based on Markov chains for the characterization of ion channels structure and dynamics. To cope with the computational challenges, the model was parallelized using a hybrid approach: cluster computing and GPGPUs (general-purpose computing on graphics processing units). Our parallel implementation of this model using a multi-GPU platform was able to reduce the execution times of the simulations from more than 6 days (on a single processor) to 21 minutes (on a small 8-node cluster equipped with 16 GPUs, i.e., 2 GPUs per node).

Implementation of a computational model of the innate immune system

In the last few years there has been an increasing interest in mathematical and computational modelling of the human immune system (HIS). Computational models of the HIS dynamics may contribute to a better understanding of the complex phenomena associate to the immune system, and support the development of new drugs and therapies for different diseases. However, the modelling of the HIS is an extremely hard task that demands huge amount of work to be performed by multi-disciplinary teams. In this scenario, the objective of this work is to model the dynamics of some cells and molecules of the HIS during an immune response to lipopolysaccharide (LPS) in a section of the tissue. The LPS constitutes the cellular wall of Gram-negative bacteria, and it is a highly immunogenic molecule, which means that it has a remarkable capacity to elicit strong immune responses.

An Electro-Mechanical Cardiac Simulator Based on Cellular Automata and Mass-Spring Models

The mechanical behavior of the heart is guided by the propagation of an electrical wave, called action potential. Many diseases have multiple effects on both electrical and mechanical cardiac physiology. To support a better understanding of the multiscale and multiphysics processes involved in physiological and pathological cardiac conditions, a lot of work has been done in developing computational tools to simulate the electro-mechanical behavior of the heart. In this work, we propose a new user-friendly and efficient tool for the electro-mechanical simulation of the cardiac tissue that is based on cellular automata and mass-spring models. The proposed tool offers a user-friendly interface that allows one to interact with the simulation on-the-fly. In addition, the simulator is parallelized with CUDA and OpenMP to further speedup the execution time of the simulations.

Computational Modeling of Microabscess Formation

Bacterial infections can be of two types: acute or chronic. The chronic bacterial infections are characterized by being a large bacterial infection and/or an infection where the bacteria grows rapidly. In these cases, the immune response is not capable of completely eliminating the infection which may lead to the formation of a pattern known as microabscess (or abscess). The microabscess is characterized by an area comprising fluids, bacteria, immune cells (mainly neutrophils), and many types of dead cells. This distinct pattern of formation can only be numerically reproduced and studied by models that capture the spatiotemporal dynamics of the human immune system (HIS). In this context, our work aims to develop and implement an initial computational model to study the process of microabscess formation during a bacterial infection.