ESTIRJ Volume 6, Issue1  Mar, 2022

Paper 1: A HYBRID LEARNING ALGORITHM FOR AUTOMATIC MRI SEGMENTATION OF NEURODEGENERATIVE DISEASES (1-10)

Abstract: Due to the relationship of various neurological illnesses with distinct regions of the brain, automated brain segmentation is an active study domain to aid medical practitioners in prognostics and diagnoses. Various technologies for automated brain segmentation have been developed using traditional methodologies such as atlas-based and pattern recognition-based methods. Deep learning approaches have recently outperformed traditional state-of-the-art methods and are gradually maturing. As a result of its ability to understand the detailed properties of high-dimensional data, deep learning has been widely used as a method for exact segmentation of brain areas. This paper proposes a network for segmenting multiple brain areas that is built on 3D convolutional neural networks and uses residual learning and dilated convolution operations to learn the end-to-end mapping from MRI volumes to voxel-level brain segments effectively. The segmentation of up to nine brain areas, including cerebrospinal fluid, white matter, and grey matter, as well as their sub-regions, is the subject of this study.

Full Text Download

Paper 2: COMPUTER AIDED COAL SEAM MODELING FOR RESOURCE ESTIMATION (11-16)

Abstract: The Badin Coal Field (BCF) is well-known for the availability of considerable coal resources. The estimation of coal resource by Bore Hole Influence Method (BHIM) is an old technique which needs to be replaced by computer aided approach for reliable estimation of natural commodity. The computer aided modeling for the quantification of coal resource has not been conducted yet, particularly for BCF, block-I. The major limitation of conventional resource estimation technique (i.e. BHIM) is a high level of uncertainty due to the ignorance of spatial continuity of the seam thickness and estimation of coal resource into tonnages instead of volume. In this study, an interpolation method named as the Inverse Distance Weighting Method (IDWM) has been used to estimate the coal resource in volume (m3) and to generate spatial distribution maps for various coal quality parameters using basic drill-hole database into the computer aided software Geovia-Surpac. The 3-Dimensional coal seam model has an estimated volume of 18.689 million m3. The spatial distribution maps of coal seam show the increase in moisture content from north-east (35-37%) toward south-west (38-39%) with a negligible variation in moisture content of coal seam. The Gross Calorific Value (GCV) is homogeneously distributed with an average GCV (3750 k. Cal/kg), except in southern block. The sulfur content of coal seam located in the south-west is less than the sulfur content from north-east. This study provides an insight for the policy makers to consider the spatial maps of coal seam for the mine-plan and operation.

Full Text Download

Paper 3: PERFORMANCE ANALYSIS OF MULTIPLE DATA TYPE DISCRETE SYSTEM THROUGH FLOW PROCESS (11-15)

Abstract: The evaluation of queueing system flow period and service process is a unique and influential concept for analyzing a flow period where every arriving data packet at any value in a system. The service and flow process evolution graphs are constructed and demonstrated in this study to calculate the time it takes for a data packet to flow in a finite capacity mechanism for queueing in discrete time with geometrically dispersed arriving data packets. We calculate the probability of totally possible preliminary states for an information packet to enter and the flow procedure to begin. We used simulation data to validate the derived analytical conclusions for probability mass function of individual beginning state also the whole probability mass function. Finally the structure of this paper in a whole scene using the 3D modelling system.

Full Text Download

p>

Paper 4: DOCUMENTATION AND MANUAL CONSERVATION STRATEGIES FOR THE STANDING STRUCTURE OF PACCA FORT HYDERABAD SINDH PAKISTAN(16-23)

Abstract: Pacca Qila is a first class monument of 18th century, the wall is disintegrated from various spots and at some huge losses are recorded. There are various factors affecting the structure of fort such as human occupants, natural and environmental effects,can further deteriorate the Pacca Fort. In order to save this monument urgent conservation works such as documentation and conservation of fort’s main gate and fortification wall needs to be done. In order to help the authorities an immediate Conservative plan and a project is required.

Full Text Download

Paper 5: ANALYSIS OF CHANNEL ESTIMATION TECHNIQUES IN MIMO-OFDM SYSTEM (24-30)

Abstract: The reception of the higher data rates, the better utilization of available bandwidth, and the reliable communication demand a better technology for the end users. The modern wireless MIMO-OFDM system is the answer for such a technology. In order to achieve these benefits, the channels of MIMO-OFDM system have to be estimated for getting the accurate channel state information. For channel estimation, different techniques like Pilot, Blind and Decision Directed based estimation techniques, are used. But the channel estimation with Pilot Symbols is selected for the purpose of analysis for this research paper. Methodology used for the analysis is such that different parameters are initialized first, and then processed on MATLAB. When the transmitter sends the data streams, it gets distorted on its way to the receiver due physical properties of the channel. Pilot symbol based channel estimation techniques like LS and MMSE are used to estimate the channel with the help of BER and SNR functions. These functions are affected by the Pilot Symbols and the Channel Taps. As the size of Pilot Symbols or Channel Taps is increased, the BER is also increased or decreased, respectively. So, it can be concluded that estimation of channels in MIMO-OFDM system is of utmost importance for effective performance of the system.

Full Text Download

Paper 6: NON-INVASIVE EEG BASED FEATURE EXTRACTION FRAMEWORK FOR MAJOR DEPRESSIVE DISORDER ANALYSIS (31-37)

Abstract: Around the world, depression and other behavioral health disorders are serious public health concerns. Persistent behavioral health issues have a wide range of consequences that affect people on a personal, relational, cultural, and social level. Major depressive disorder (MDD) is a mental illness that affects people of all ages around the world. It has grown into a major global health issue as well as an economic burden. Clinicians have traditionally used a variety of medications to limit the growth of this disease at an early stage. The goal of this research is to improve depression diagnosis by altering electroencephalogram (EEG) signals and extracting the Differential Entropy (DE) and Power Spectral Density (PSD), using machine learning and deep learning techniques. This study analyzed the EEG signals of 30 healthy people and 34 people with Major Depressive Disorder (MDD). K-nearest neighbors (KNN) had the highest accuracy among machine learning algorithms of 99%, while Support vector machine (SVM) had achieved 95% accuracy. The developed Deep Learning approach, convolutional neural network (CNN), achieved 99% accuracy. With these promising results, this study establishes the viability of EEG-based diagnosis of MDD.

Full Text Download

Paper 6: FLOW PROCESS ANALYSIS OF TWO TYPE DATA HAVING PRE-EMPTIVE PRIORITY TECHNIQUE (0-000)

Abstract: aaaaaaaaaaaaaaa

Full Text Download