International Journal of Biology and Biomedical Engineering


ISSN: 1998-4510
Volume 13, 2019

Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of NAUN Journals is adapted to the 'continuously updated' model. What this means is that instead of being separated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.

Main Page

Submit a paper | Submission terms | Paper format

Special Issue: Fifth International Conference on Parallel, Distributed and Grid Computing

 


Volume 13, 2019


Title of the Paper: Variable Complexity Neural Networks Comparison for Pollen Classification

 

Authors: Aysha Kadaikar, Yan Pan, Qiaoxi Zhang, Patricia Conde-Cespedes, Maria Trocan, Frederic Amiel, Benjamin Guinot

Pages: 38-45

Abstract: This paper deals with the problem of classifying the pollen grains observed in a microscope view acquired by a collector of ambient air particles. This classification is usually performed by a highly skilled human operator observing the microscope slide to detect the presence of pollen grains, count them and sort them according to their taxa. However these tasks become particularly heavy in the mid-season because of the huge quantity of pollen produced. This paper compares the use of three neural networks (NN) to classify the pollen grains observed which are a modified version of LeNet5, ResNet50 and AlexNet. The first two have been conceived more for non-natural images and the last one for natural images. Simulation shows that ResNet50 and AlexNet particularly lead to good performance in terms of accuracy for this kind of images. AlexNet is finally a good compromise for pollen classification when adding a constraint on the computational complexity.


Title of the Paper: SOA based Multi-Agent Approach for Biological Data Searching and Integration

 

Authors: Veska Gancheva

Pages: 32-37

Abstract: Major challenge in the analysis of biological data is to propose an integrated and modern access to the progressively increasing amounts of data in multiple formats, and efficient approaches for their processing. Models for extraction and integration of large amount of genomics data, as well as problems related to heterogeneity, distribution and compatibility of data are presented in this paper. SOA based multi-agent approach for biological data searching and integration is proposed. A conceptual architecture for integrating of distributed biological data based on SOA is designed. The architecture is aimed to automate the data integration and allows the rapid management of large volumes of diverse data sets represented in different formats - relational, NoSQL, flat files. The integration of different databases is solved by using multi-agent architecture. The integration system consists of services for transforming the common request into a specific language request for each local database, depending on its type. The conceptual database integration is solved by applying translating query approach. Each integrated database is represented by a separate conceptual scheme called a virtual scheme. This scheme is generated in the collating process, which compares structural elements from the database to the conceptual model. Service oriented multi-agent system for searching of biological data from different sources that sends queries to multiple databases and then compiles the results into a list, depending on the type of source is developed. The system allows the user to set search criteria and access multiple databases simultaneously. The services allow the system to be accessed over the Internet by multiple clients (mobile phones, web browsers, desktop applications) and serving a wide range of users simultaneously.


Title of the Paper: Molecular Dynamics Simulation of Interaction of Lysine Dendrigraft of 2nd Generation with Stack of Amyloid Peptides

 

Authors: I. Neelov, D. Khamidova, V. Bezrodnyi, S. Mikhtaniuk

Pages: 26-31

Abstract: In present paper, molecular dynamics simulation is used to study destruction of stack of short amyloid peptide molecules by oppositely charged dendrigraft of 2nd generation. Dendrimers and dendrigraftss are often used in biomedicine for delivery of drugs and other biological molecules. They also could be used as antibacterial, antiviral and antiamyloid agents. Since lysine dendrimers and dendrigrafts are less toxic than many other conventional synthetic dendrimers they were chosen for present study and two systems consisting of 2nd generation dendrigraft and stack of 8 or 16 short amyloid peptide molecules were simulated by the method of molecular dynamics in water. It was demonstrated that lysine dendrigraft destroys both studied amyloid stacks and forms stable complexes with their peptide molecules. The final structures of the complexes in equilibrium state were studied also. It was shown that peptides in complexes stay mainly on the surface of dendrigrafts and do not penetrate into them. The results obtained in present paper could be useful for elaboration in future the anti-amyloid agents for treatment of Alzheimer's disease, since it is believed that one of the sources of this disease is the formation of toxic amyloid oligomers and fibrils.


Title of the Paper: On the Use of Anatomage Table as Diagnostic Tool

 

Authors: Alexandre Taoum, Rihab Sadqi, Mustapha Zidi, Alexandra d’Anglemont de Tassigny, Kawtar Megdiche, Nabil Ngote

Pages: 20-25

Abstract: Anatomy is one of the most important course in medical curriculum. To be effective, medical student must do several hours of practice on human cadavers. However, today, less and less corpses are available to be dissected. Therefore, Anatomage has developed a virtual dissection table known as Anatomage Table. In addition to filling the lack of corpses and being a very useful tool in anatomy classes, this table can open pathological images. The main purpose of this article is to show the investigation carried out to be able to use the table as a diagnostic tool. Before opening DICOM image into the table, Matlab was used to make the files readable by the Table. This paper deals with the opening of patients pathological images from differents type of medical imaging modalities. Indeed, the opening of pathological images turns the Anatomage Table into a very relevant tool in the diagnosis of several pathologies. Experimental 3D reconstruction of pathological patients are presented in order to show the effectiveness of the proposed method. After opening pathological patient files, some hard tissue and soft tissue pathologies were diagnosed thanks to the use of Anatomage Table. That’s why the obtained results are largely satisfactory, indicating a promising use in the healing process.


Title of the Paper: Light Harvesting Complex LH2 - Simulation of Spectral Profiles

 

Authors: Pavel HeĊ™man, David Zapletal

Pages: 10-19

Abstract: Absorption and steady state fluorescence spectra are simulated for relatively simple circular molecular system. It can be treated as a model of peripheral light harvesting complex LH2 from purple bacterium Rhodopseudomonas acidophila. Both rings (B850 ring and B800 one) are included in our model. The spectra are calculated within full Hamiltonian model and compared for two types of slow fluctuations. Gaussian fluctuations in local excitations energies and Gaussian fluctuations in positions of bacteriochlorophylls are considered. Fast fluctuations, interaction with phonon bath, is also taking into account for low and room temperature. The resulting spectra show strong dependence on temperature. Splitting of both spectra are visible at low temperature. The differences caused by static disorder type are also remarkable. Localization of exciton states contributing to the steady state fluorescence spectra is investigated by thermally averaged by participation ratio.


Title of the Paper: An Iterative Map Analysis of Heart Rate Variability in the Neighborhood of Stress Induced Chaos

 

Authors: Theophilus A. Fashanu, Olagoke K. Ale, Ayo Oyediran, Viyon Dansu

Pages: 1-9

Abstract: This work examine the viability of one-dimensional Poincare section as an iterative tool for the inversion of electrocardiogram obtained from stressed cardiovascular system. The response of Heart Rate Variability to induced stress in a healthy subject on a Modified Bruce Treadmill Protocol was processed to derive R-peaks in six batches of 5.06 minutes per batch. For each stage, a corresponding iterative map was constructed with the sequenced R-peaks using the Kubios software. With increasing intensity of induced stress, the maps outline the nonlinear dynamic response of the subject’s autonomous nervous system. At the combined warm up stages one and two of the protocol, 1.6% offshoot in ANS activities is indicated with higher rate of parasympathetic activities. Response to low intensity stress condition showed symmetric ANS activities. However, with high intensity stress at the fifth stage, the map indicated an exponential increase in the number of heart beat orbits. However, the increment is characterized by skewed ANS activities with 15% acceleration of sympathetic ANS and 5% deceleration of parasympathetic activities. This profile is preserved at the final stage. In effect, the topology of the iterative maps of stages five and six of the protocol shows an inner ANS dynamical structure that is characterized by positive Lyapunov exponent. This limiting dynamical response determines the cardio-pulmonary reserve of the subject. Thus, Poincare iterative maps provide high resolution computational tool for identifying the onset of chaotic motion in stressed cardiovascular systems.