High Performance GP-Based Approach for fMRI Big Data Classification Proceedings of the Practice and Experience in Advanced Research Computing [PEARC17] , ACM, July 2017 Other Authors: Amir H. Gandomi, Anke Meyer-Baese We consider resting-state Functional Magnetic Resonance Imaging (fMRI) of two classes of patients: one that took the drug N-acetylcysteine (NAC) and the other one a placebo before and after a smoking cessation treatment. Our goal was to classify the relapse in nicotine-dependent patients as treatment or non-treatment based on their fMRI scans. 80% accuracy was obtained using Independent Component Analysis (ICA) along with Genetic Programming (GP) classifier using High Performance Computing (HPC) which we consider significant enough to suggest that there is a difference in the resting-state fMRI images of a smoker that undergoes this smoking cessation treatment compared to a smoker that receives a placebo.
Building energy consumption forecast using multi-objective genetic programming Journaleee of Measurement , Elsevier, January 2018 Other Author: Amir H. Gandomi A multi-objective genetic programming (MOGP) technique with multiple genes is proposed to formulate the energy performance of residential buildings. Here, it is assumed that loads have linear relation in terms of genes. On this basis, an equation is developed by MOGP method to predict both heating and cooling loads. The proposed evolutionary approach optimizes the most significant predictor input variables in the model for both accuracy and complexity, while simultaneously solving the unknown parameters of the model. In the proposed energy performance model, relative compactness has the most and orientation the least contribution. The proposed MOGP model is simple and has a high degree of accuracy. The results show that MOGP is a suitable tool to generate solid models for complex nonlinear systems with capability of solving big data problems via parallel algorithms.
High Performance GP-Based Approach for fMRI Big Data Classification Proceedings of the Practice and Experience in Advanced Research Computing [PEARC17] , ACM, July 2017 Other Authors: Amir H. Gandomi, Anke Meyer-Baese We consider resting-state Functional Magnetic Resonance Imaging (fMRI) of two classes of patients: one that took the drug N-acetylcysteine (NAC) and the other one a placebo before and after a smoking cessation treatment. Our goal was to classify the relapse in nicotine-dependent patients as treatment or non-treatment based on their fMRI scans. 80% accuracy was obtained using Independent Component Analysis (ICA) along with Genetic Programming (GP) classifier using High Performance Computing (HPC) which we consider significant enough to suggest that there is a difference in the resting-state fMRI images of a smoker that undergoes this smoking cessation treatment compared to a smoker that receives a placebo.
fMRI Smoking Cessation Classification Using Genetic Programming Data Science meets Optimization (DSO), May 2017 Other Authors: Amir H. Gandomi, Ian McCaan, Anneke Goudriaan, Lianne Schmaal, Anke Meyer-Baese Resting-state functional magnetic resonance imaging (fMRI) images allow us to see the level of activity in a patient's brain. We consider fMRI of patients before and after they underwent a smoking cessation treatment. Two classes of patients have been studied here, that one took the drug N-acetylcysteine and the ones took a placebo. Our goal was to classify the relapse in nicotine-dependent patients as treatment or non-treatment based on their fMRI scans. The image slices of brain are used as the variable and as results here we deal with a big data problem with about 240,000 inputs. To handle this problem, the data had to be reduced and the first process in doing that was to create a mask to apply to all images. The mask was created by averaging the before images for all patients and selecting the top 40% of voxels from that average. This mask was then applied to all fMRI images for all patients. The average of the difference in the before treatment and after fMRI images for each patient were found and these were flattened to one dimension. Then a matrix was made by stacking these 1D arrays on top of each other and a data reduction algorithm was applied on it. Lastly, this matrix was fed into some machine learning and Genetic Programming algorithms and leave-one-out cross-validation was used to test the accuracy. Out of all the data reduction machine learning algorithms used, the best accuracy was obtained using Principal Component Analysis along with Genetic Programming classifier. This gave an accuracy of 74%, which we consider significant enough to suggest that there is a difference in the resting-state fMRI images of a smoker that undergoes this smoking cessation treatment compared to a smoker that receives a placebo.
Reconfigurable wearable to monitor physiological variables and movement Proc. SPIE Vol. 10216, Smart Biomedical and Physiological Sensor Technology XIV, April 2017 Other Authors: Antonio Garcia, Diego Morales, Anke Meyer-Baese This article presents a preliminary prototype of a wearable instrument for oxygen saturation and ECG monitoring. The proposed measuring system is based on the light reflection variability of a LED emission on the subject temple. Besides, the system has the capacity to incorporate electrodes to obtain ECG measurements. All measurements are stored and transmitted to a mobile device (tablet or smartphone) through a Bluetooth link.
Dynamical graph theory networks techniques for the analysis of sparse connectivity networks in dementia Proc. SPIE Vol. 10216, Smart Biomedical and Physiological Sensor Technology XIV, April 2017 Other Authors: Katja Pinker, Anke Meyer-Baese Graph network models in dementia have become an important computational technique in neuroscience to study fundamental organizational principles of brain structure and function of neurodegenerative diseases such as dementia. The graph connectivity is reflected in the connectome, the complete set of structural and functional connections of the graph network, which is mostly based on simple Pearson correlation links. In contrast to simple Pearson correlation networks, the partial correlations (PC) only identify direct correlations while indirect associations are eliminated. In addition to this, the state-of-the-art techniques in brain research are based on static graph theory, which is unable to capture the dynamic behavior of the brain connectivity, as it alters with disease evolution. We propose a new research avenue in neuroimaging connectomics based on combining dynamic graph network theory and modeling strategies at different time scales. We present the theoretical framework for area aggregation and time-scale modeling in brain networks as they pertain to disease evolution in dementia. This novel paradigm is extremely powerful, since we can derive both static parameters pertaining to node and area parameters, as well as dynamic parameters, such as system’s eigenvalues. By implementing and analyzing dynamically both disease driven PC-networks and regular concentration networks, we reveal differences in the structure of these network that play an important role in the temporal evolution of this disease. The described research is key to advance biomedical research on novel disease prediction trajectories and dementia therapies.
The driving regulators of the connectivity protein network of brain malignancies Proc. SPIE Vol. 10216, Smart Biomedical and Physiological Sensor Technology XIV, April 2017 Other Authors: Katja Pinker, Anke Meyer-Baese An important problem in modern therapeutics at the proteomic level remains to identify therapeutic targets in a plentitude of high-throughput data from experiments relevant to a variety of diseases. This paper presents the application of novel modern control concepts, such as pinning controllability and observability applied to the glioma cancer stem cells (GSCs) protein graph network with known and novel association to glioblastoma (GBM). The theoretical frameworks provides us with the minimal number of "driver nodes", which are necessary, and their location to determine the full control over the obtained graph network in order to provide a change in the network’s dynamics from an initial state (disease) to a desired state (non-disease). The achieved results will provide biochemists with techniques to identify more metabolic regions and biological pathways for complex diseases, to design and test novel therapeutic solutions.
Characteristics of Nano-Structures SciComp Conference, August 2016 As a final project of the Scientific Communications class, a Poster plus ten minutes talk under supervision of Dr. John Burkardt has been preseneted.
Modeling the Zombie Apocalypse XSEDE Competition, July 2016 Other Team Members: Brian Bartoldson, Eitan Lees, Alex Townsend, Ian McCann We simulated a model for the zombie apocalypse on an island where humans can meet zombies and interact by being bitten and becoming a zombie, being killed, or killing the zombie.
Pharmecokinetic Model of Drug Dosage and Concentrations XSEDE Competition, April 2016 Other Team Members: Brian Bartoldson, Eitan Lees, Alex Townsend, Ian McCann In order to be effective, the concentration of a drug in the bloodstream needs to reach a medicinal level.Below this level, the drug will be ineffective. For some drugs, there is also the limitation that above some concentration level they become toxic. Thus, it is critical to vary the dosage such that the steady state concentration is between the effective and toxic dose.
We have coded the aforementioned model of drug dosage and concentrations in Python.
That was for XSEDE 2016 competition at Florida State University.
in order to get a better understanding of the unique properties and phenomena
of nano-fluidics. The previous modeling and simulation efforts were based
on diffusion of atoms or molecules that were thrown to the nanotubes with initial
velocities. This talk has shed some light on the flow of fluids using molecular
dynamic simulations of different types of carbon nanotubes that were embedded
in liquid argon using a moving wall piston of graphene. We focused on analyzing
pressure difference, velocities, and momentum conservation in different regions.
Fluid Flow Through Carbon Nanotubes And Graphene Based Nanostructures OhioLINK ETD, August 2015 Abstract: The investigation into the behavior of the fluids in nanoscale channels,
such as carbon nanotubes leads us to a new approach in the field of nanoscience. This
is referred to as nano-fluidics, which can be used in nano-scale filtering and as
nano-pipes for conveying fluids. The behavior of fluids in nano-fluidic devices is
very different from the corresponding behavior in microscopic and macroscopic channels.
In this study, we investigate the fluid flow through carbon nanotubes and graphene
based nanostructures using a molecular dynamics (MD) method at a constant
temperature.Three different models were created which contain single-walled carbon
nanotube, graphene, and a combination of both. Liquid argon is used as fluid in the
system. In the previous investigations, they were considered bombarding the atoms
towards the carbon nanotubes like bullets from a gun, and due to the interactions,
they lost most of their momentum. Thus, the chance for the atoms to pass through
the carbon nanotube was very low. Here, we employed a new approach using a moving
graphene wall to push the argon fluid towards the confinements of the systems.
By performing this method, we have tried to make a continuum flow to find out
how the physical quantities such as, position, velocity, pressure, and energy
change when the fluid flow reaches the confinements of the systems.