Search results for: histogram normalization

Authors: H. Abas Nejad, A. R. Teymoori

Abstract:

Facial expression recognition is one of the open problems in computer vision. Robust neutral face recognition in real time is a major challenge for various supervised learning based facial expression recognition methods. This is due to the fact that supervised methods cannot accommodate all appearance variability across the faces with respect to race, pose, lighting, facial biases, etc. in the limited amount of training data. Moreover, processing each and every frame to classify emotions is not required, as the user stays neutral for the majority of the time in usual applications like video chat or photo album/web browsing. Detecting neutral state at an early stage, thereby bypassing those frames from emotion classification would save the computational power. In this work, we propose a light-weight neutral vs. emotion classification engine, which acts as a preprocessor to the traditional supervised emotion classification approaches. It dynamically learns neutral appearance at Key Emotion (KE) points using a textural statistical model, constructed by a set of reference neutral frames for each user. The proposed method is made robust to various types of user head motions by accounting for affine distortions based on a textural statistical model. Robustness to dynamic shift of KE points is achieved by evaluating the similarities on a subset of neighborhood patches around each KE point using the prior information regarding the directionality of specific facial action units acting on the respective KE point. The proposed method, as a result, improves ER accuracy and simultaneously reduces the computational complexity of ER system, as validated on multiple databases.

Keywords: neutral vs. emotion classification, Constrained Local Model, procrustes analysis, Local Binary Pattern Histogram, statistical model

Procedia PDF Downloads 319

Authors: Yildirim Ortaoglan

Abstract:

The attempt to make thought school-like by fitting it into various frameworks with the institutionalization of it is almost simultaneous with philosophy itself. What once sprouted in the “academia” of old has institutionalized under the enlightenment's light, becoming the fundamental space reflecting the spirit of its age. However, the shift from the thinking temple where truth's knowledge was sought to functional spaces where power/power relations are constructed indicates a significant rupture in the meaning of school. Therefore, a genealogical inquiry into the meaning of the school can provide us with a path toward understanding how it should be approached in contemporary times. From this perspective, it is essential to highlight how power/power relations operate in the school in terms of disciplinary practices, temporal management, and spatial organization to construct a distinct subjectivation. Recognizing that the changing and evolving nature of education is related to the structure of space can be understood by revealing how disciplinary power and bio-power, two fundamental aspects of genealogical research, operate. In disciplinary power, the relationship of the subject with discipline, temporal management, and space is about improvement and normalization, while in biopower, it manifests in maximizing utility, increasing free time, and constructing spaces that seem more vital. These indicators not only facilitate the formation of students as a subjectivation but also enable the condition of the possibility of power/power relations. Because power is not applied to subjects but used by them for passage, and behind this lies the idea that the individual is already one of the components of power. As one of the components of power, in terms of subjectivation type, the student is one of the primary targets of power relations. Therefore, conducting a genealogical inquiry of the student as a type of subjectivation and the school as its living area from the philosophical foundations of education may offer a new opportunity for thinking about the contemporary crisis of thought. Within the framework of this possibility, our investigation will consider which aspects of the school and the student, brought together for educational purposes, can be thought of within and beyond power/power relations.

Keywords: power, education, space, school, student, discipline

Procedia PDF Downloads 22

Authors: Shraddha Srivastava, N. K. Painuly, S. P. Mishra, Navin Singh, Muhsin Punchankandy, Kirti Srivastava, M. L. B. Bhatt

Abstract:

Various optimization methods used in interstitial brachytherapy are based on dwell positions and dwell weights alteration to produce dose distribution based on the implant geometry. Since these optimization schemes are not anatomy based, they could lead to deviations from the desired plan. This study was henceforth carried out to compare anatomy-based Inverse Planning Simulated Annealing (IPSA) optimization technique with graphical and geometrical optimization methods in interstitial high dose rate brachytherapy planning of cervical carcinoma. Six patients with 12 CT data sets of MUPIT implants in HDR brachytherapy of cervical cancer were prospectively studied. HR-CTV and organs at risk (OARs) were contoured in Oncentra treatment planning system (TPS) using GYN GEC-ESTRO guidelines on cervical carcinoma. Three sets of plans were generated for each fraction using IPSA, graphical optimization (GrOPT) and geometrical optimization (GOPT) methods. All patients were treated to a dose of 20 Gy in 2 fractions. The main objective was to cover at least 95% of HR-CTV with 100% of the prescribed dose (V100 ≥ 95% of HR-CTV). IPSA, GrOPT, and GOPT based plans were compared in terms of target coverage, OAR doses, homogeneity index (HI) and conformity index (COIN) using dose-volume histogram (DVH). Target volume coverage (mean V100) was found to be 93.980.87%, 91.341.02% and 85.052.84% for IPSA, GrOPT and GOPT plans respectively. Mean D90 (minimum dose received by 90% of HR-CTV) values for IPSA, GrOPT and GOPT plans were 10.19 ± 1.07 Gy, 10.17 ± 0.12 Gy and 7.99 ± 1.0 Gy respectively, while D100 (minimum dose received by 100% volume of HR-CTV) for IPSA, GrOPT and GOPT plans was 6.55 ± 0.85 Gy, 6.55 ± 0.65 Gy, 4.73 ± 0.14 Gy respectively. IPSA plans resulted in lower doses to the bladder (D₂

Keywords: cervical cancer, HDR brachytherapy, IPSA, MUPIT

Procedia PDF Downloads 159

Authors: Hazel Ann Gianelli Cu, Stephanie Co, Radcliff Cobankiat

Abstract:

Objectives: TRC101/Veverimer is an orally administered, non absorbed, sodium- and counterion-free hydrochloric acid binder for the treatment of metabolic acidosis associated with chronic kidney disease. The main objective of this study is to determine the efficacy of TRC 101/ Veverimer 6g/day in increasing serum bicarbonate levels of chronic kidney disease patients with metabolic acidosis. In this meta analysis, we also aim to look at safety outcomes, adverse effects and if the level of serum bicarbonate reached metabolic alkalosis when given TRC101/Veverimer. Methodology: Pubmed, Cochrane, Google Scholar and Science direct were used to search for randomized controlled trials about TRC101/Veverimer use in Chronic kidney disease patients with metabolic acidosis. Search strategy according to the Prisma checklist was done with evaluation of biases and synthesis of results using the Cochrane Review Manager software 5.4. Results: Two randomized controlled trials involving 371 chronic kidney disease patients were included in this study. Results show there was a significant increase in the serum bicarbonate level when given TRC101/Veverimer compared to the placebo. Both studies had a significant number of participants who completed the studies until the end. P value of <0.00001 was used in both studies with a confidence interval of 95%. Conclusion: TRC101/Veverimer 6g/day was shown to effectively and safely increase serum bicarbonate or achieve normalization in chronic kidney disease patients with metabolic acidosis as compared with a placebo. This was associated with delayed progression of kidney disease with improvement of physical functioning, however longer duration of future studies is ideal in order to assess further the long advantages and consequences of TRC 101/Veverimer.

Keywords: chronic kidney disease, metabolic acidosis, Veverimer, TRC101

Procedia PDF Downloads 170

Authors: Bernardo C. P. Albuquerque, Darym J. F. Campos

Abstract:

Increasing our ability to solve complex engineering problems is directly related to the processing capacity of computers. By means of such equipments, one is able to fast and accurately run numerical algorithms. Besides the increasing interest in numerical simulations, probabilistic approaches are also of great importance. This way, statistical tools have shown their relevance to the modelling of practical engineering problems. In general, statistical approaches to such problems consider that the random variables involved follow a normal distribution. This assumption tends to provide incorrect results when skew data is present since normal distributions are symmetric about their means. Thus, in order to visualize and quantify this aspect, 9 statistical distributions (symmetric and skew) have been considered to model a hypothetical slope stability problem. The data modeled is the friction angle of a superficial soil in Brasilia, Brazil. Despite the apparent universality, the normal distribution did not qualify as the best fit. In the present effort, data obtained in consolidated-drained triaxial tests and saturated direct shear tests have been modeled and used to analytically derive the probability density function (PDF) of the safety factor of a hypothetical slope based on Mohr-Coulomb rupture criterion. Therefore, based on this analysis, it is possible to explicitly derive the failure probability considering the friction angle as a random variable. Furthermore, it is possible to compare the stability analysis when the friction angle is modelled as a Dagum distribution (distribution that presented the best fit to the histogram) and as a Normal distribution. This comparison leads to relevant differences when analyzed in light of the risk management.

Keywords: statistical slope stability analysis, skew distributions, probability of failure, functions of random variables

Procedia PDF Downloads 309

Authors: Hyeon-Kyo Lim, Hyunjung Kim, Kune-Woo Lee

Abstract:

Decommissioning process of nuclear facilities can be said to consist of a sequence of problem solving activities, partly because there may exist working environments contaminated by radiological exposure, and partly because there may also exist industrial hazards such as fire, explosions, toxic materials, and electrical and physical hazards. As for an individual hazard factor, risk assessment techniques are getting known to industrial workers with advance of safety technology, but the way how to integrate those results is not. Furthermore, there are few workers who experienced decommissioning operations a lot in the past. Therefore, not a few countries in the world have been trying to develop appropriate counter techniques in order to guarantee safety and efficiency of the process. In spite of that, there still exists neither domestic nor international standard since nuclear facilities are too diverse and unique. In the consequence, it is quite inevitable to imagine and assess the whole risk in the situation anticipated one by one. This paper aimed to find out an appropriate technique to integrate individual risk assessment results from the viewpoint of experts. Thus, on one hand the whole risk assessment activity for decommissioning operations was modeled as a sequence of individual risk assessment steps, and on the other, a hierarchical risk structure was developed. Then, risk assessment procedure that can elicit individual hazard factors one by one were introduced with reference to the standard operation procedure (SOP) and hierarchical task analysis (HTA). With an assumption of quantification and normalization of individual risks, a technique to estimate relative weight factors was tried by using the conventional Analytic Hierarchical Process (AHP) and its result was reviewed with reference to judgment of experts. Besides, taking the ambiguity of human judgment into consideration, debates based upon fuzzy inference was added with a mathematical case study.

Keywords: decommissioning, risk assessment, analytic hierarchical process (AHP), fuzzy inference

Procedia PDF Downloads 400

Authors: V. Veeraprathap, G. S. Harish, G. Narendra Kumar

Abstract:

Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.

Keywords: artificial neural networks, ANN, discrete wavelet transform, DWT, gray-level co-occurrence matrix, GLCM, k-nearest neighbor, KNN, region of interest, ROI

Procedia PDF Downloads 124

Authors: A. Farhad Sadri, B. Ali Farhadi, C. Nasim Shalamzari

Abstract:

Waste management is a principle of urban and environmental governance. Waste management in Tehran metropolitan requires good strategies for better governance. Using of good urban governance principles together with eight main indexes can be an appropriate base for this aim. One of the reasonable tools in this field is usage of SWOT methods which provides possibility of comparing the opportunities, threats, weaknesses, and strengths by using IFE and EFE matrixes. The results of the above matrixes, respectively 2.533 and 2.403, show that management system of Tehran metropolitan wastes has performed weak regarding to internal factors and has not have good performance regarding using the opportunities and dealing with threats. In this research, prioritizing and describing the real value of each 24 strategies in waste management in Tehran metropolitan have been surveyed considering good governance derived from Quantitative Strategic Planning Management (QSPM) by using Kolomogrof-Smirnoff by 1.549 and significance level of 0.073 in order to define normalization of final values and all of the strategies utilities and Variance Analysis of ANOVA has been calculated for all SWOT strategies. Duncan’s test results regarding four WT, ST, WO, and SO strategies show no significant difference. In addition to mean comparison by Duncan method in this research, LSD (Lowest Significant Difference test) has been used by probability of 5% and finally, 7 strategies and final model of Tehran metropolitan waste management strategy have been defined. Increasing the confidence of people with transparency of budget, developing and improving the legal structure (rule-oriented and law governance, more responsibility about requirements of private sectors, increasing recycling rates and real effective participation of people and NGOs to improve waste management (contribution) and etc, are main available strategies which have been achieved based on good urban governance management principles.

Keywords: waste, strategy, environmental management, urban good governance, SWOT

Procedia PDF Downloads 292

Authors: M. R. Ahmad, I. Musirin, M. M. Othman, N. A. Rahmat

Abstract:

Plug-in Hybrid-Electric Vehicle (PHEV) has gained immense popularity in recent years. PHEV offers numerous advantages compared to the conventional internal-combustion engine (ICE) vehicle. Millions of PHEVs are estimated to be on the road in the USA by 2020. Uncoordinated PHEV charging is believed to cause severe impacts to the power grid; i.e. feeders, lines and transformers overload and voltage drop. Nevertheless, improper PHEV data model used in such studies may cause the findings of their works is in appropriated. Although smart charging is more attractive to researchers in recent years, its implementation is not yet attainable on the street due to its requirement for physical infrastructure readiness and technology advancement. As the first step, it is finest to study the impact of charging PHEV based on real vehicle travel data from National Household Travel Survey (NHTS) and at present charging rate. Due to the lack of charging station on the street at the moment, charging PHEV at home is the best option and has been considered in this work. This paper proposed a technique that comprehensively presents the impact of charging PHEV on power system networks considering huge numbers of PHEV samples with its traveling data pattern. Vehicles Charging Load Profile (VCLP) is developed and implemented in IEEE 30-bus test system that represents a portion of American Electric Power System (Midwestern US). Normalization technique is used to correspond to real time loads at all buses. Results from the study indicated that charging PHEV using opportunity charging will have significant impacts on power system networks, especially whereas bigger battery capacity (kWh) is used as well as for higher penetration level.

Keywords: plug-in hybrid electric vehicle, transportation electrification, impact of charging PHEV, electricity demand profile, load profile

Procedia PDF Downloads 252

Authors: Waqqas-ur-Rehman Butt, Martin Servin, Marion Pause

Abstract:

In recent years, object detection has gained much attention and very encouraging research area in the field of computer vision. The robust object boundaries detection in an image is demanded in numerous applications of human computer interaction and automated surveillance systems. Many methods and approaches have been developed for automatic object detection in various fields, such as automotive, quality control management and environmental services. Inappropriately, to the best of our knowledge, object detection under illumination with shadow consideration has not been well solved yet. Furthermore, this problem is also one of the major hurdles to keeping an object detection method from the practical applications. This paper presents an approach to automatic object detection in images under non-standardized environmental conditions. A key challenge is how to detect the object, particularly under uneven illumination conditions. Image capturing conditions the algorithms need to consider a variety of possible environmental factors as the colour information, lightening and shadows varies from image to image. Existing methods mostly failed to produce the appropriate result due to variation in colour information, lightening effects, threshold specifications, histogram dependencies and colour ranges. To overcome these limitations we propose an object detection algorithm, with pre-processing methods, to reduce the interference caused by shadow and illumination effects without fixed parameters. We use the Y CrCb colour model without any specific colour ranges and predefined threshold values. The segmented object regions are further classified using morphological operations (Erosion and Dilation) and contours. Proposed approach applied on a large image data set acquired under various environmental conditions for wood stack detection. Experiments show the promising result of the proposed approach in comparison with existing methods.

Keywords: image processing, illumination equalization, shadow filtering, object detection

Procedia PDF Downloads 192

Authors: Hadjer Sadoune, Liza Lamini, Scherazade Krim, Amel Djouadi, Rachida Rihani

Abstract:

Deep neural networks are highly regarded for their accuracy in predicting intricate fermentation processes. Their ability to learn from a large amount of datasets through artificial intelligence makes them particularly effective models. The primary obstacle in improving the performance of these models is to carefully choose the suitable hyperparameters, including the neural network architecture (number of hidden layers and hidden units), activation function, optimizer, learning rate, and other relevant factors. This study predicts biogas production from real wastewater treatment plant data using a sophisticated approach: hybrid Bayesian optimization with a tree-structured Parzen estimator (BO-TPE) for an optimised deep neural network (DNN) model. The plant utilizes an Upflow Anaerobic Sludge Blanket (UASB) digester that treats industrial wastewater from soft drinks and breweries. The digester has a working volume of 1574 m3 and a total volume of 1914 m3. Its internal diameter and height were 19 and 7.14 m, respectively. The data preprocessing was conducted with meticulous attention to preserving data quality while avoiding data reduction. Three normalization techniques were applied to the pre-processed data (MinMaxScaler, RobustScaler and StandardScaler) and compared with the Non-Normalized data. The RobustScaler approach has strong predictive ability for estimating the volume of biogas produced. The highest predicted biogas volume was 2236.105 Nm³/d, with coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE) values of 0.712, 164.610, and 223.429, respectively.

Keywords: anaerobic digestion, biogas production, deep neural network, hybrid bo-tpe, hyperparameters tuning

Procedia PDF Downloads 13

Authors: Hong Beom Hur

Abstract:

Politics in America is polarized. The best illustration of this is the 2016 presidential election result map. States with megacities like California, New York, Illinois, Virginia, and others are marked blue to signify the color of the Democratic party. States located in inland and south like Texas, Florida, Tennesse, Kansas and others are marked red to signify the color of the Republican party. Such a stark difference between two colors, red and blue, combined with geolocations of each state with their borderline remarks one central message; America is divided into two colors between urban Democrats and rural Republicans. This paper seeks to defy the visualization by pointing out its limitations and search for alternative ways to visualize the 2016 election result. One such limitation is that geolocations of each state and state borderlines limit the visualization of population density. As a result, the election result map does not convey the fact that Clinton won the popular vote and only accentuates the voting patterns of urban and rural states. The paper seeks whether an alternative narrative can be observed by factoring in the population number into the size of each state and manipulating the state borderline according to the normalization. Yet another alternative narrative may be reached by factoring the size of each state by the number of the electoral college of each state by voting and visualize the number. Other alternatives will be discussed but are not implemented in visualization. Such methods include dividing the land of America into about 120 million cubes each representing a voter or by the number of whole population 300 million cubes. By exploring these alternative methods to visualize the politics of the 2016 election map, the public may be able to question whether it is possible to be free from the narrative of the divide-conquer when interpreting the election map and to look at both parties as a story of the United States of America.

Keywords: 2016 U.S. presidential election, data visualization, population scale, geo-political

Procedia PDF Downloads 96

Authors: Sajjad Shojaeidoust, Mohsen Ghanbarzadeh, Abdolhamid Habibi

Abstract:

Exercise-induced bronchospasm (EIB) is a transitory condition of airflow obstruction that is associated with physical activities. It is noted that high ventilation can lead to an increase in the heat and reduce in the moisture in airways resistance of trachea. Also causes of pathophysiological mechanism are EIB. Accordingly, studying some parameters of pulmonary function (FVC, FEV1) among active people seems quintessential. The aim of this study was to analyze bronchospasm following the implementation of a program of maximum aerobic exercise in active men at Chamran University of Ahwaz. Method: In this quasi-experimental study, the population consisted of all students at Chamran University. Among from 55 participants, of which, 15 were randomly selected as the experimental group. In this study, the size of the maximum oxygen consumption was initially measured, and then, based on the maximum oxygen consumed, the active individuals were identified. After five minutes’ warm-up, Strand treadmill exercise test was taken (one session) and pulmonary parameters were measured at both pre- and post-tests (spirometer). After data normalization using KS and non-normality of the data, the Wilcoxon test was used to analyze the data. The significance level for all statistical surveys was considered p≤0/05. Results: The results showed that the ventilation factors and bronchospasm (FVC, FEV1) in the pre-test and post-test resulted in no significant difference among the active people (p≥0/05). Discussion and conclusion: Based on the results observed in this study, it appears that pulmonary indices in active individuals increased after aerobic test. The increase in this indicator in active people is due to increased volume and elasticity of the lungs as well. In other words, pulmonary index is affected by rib muscles. It is considered that progress over respiratory muscle strength and endurance has raised FEV1 in the active cases.

Keywords: aerobic active maximum, bronchospasm, pulmonary function, spirometer

Procedia PDF Downloads 260

Authors: Nkechinyere Joy Olawuyi, Babajide Samuel Afolabi, Bola Ibitoye

Abstract:

This study collected a preprocessed dataset of chest radiographs and formulated a deep neural network model for detecting abnormalities. It also evaluated the performance of the formulated model and implemented a prototype of the formulated model. This was with the view to developing a deep neural network model to automatically classify abnormalities in chest radiographs. In order to achieve the overall purpose of this research, a large set of chest x-ray images were sourced for and collected from the CheXpert dataset, which is an online repository of annotated chest radiographs compiled by the Machine Learning Research Group, Stanford University. The chest radiographs were preprocessed into a format that can be fed into a deep neural network. The preprocessing techniques used were standardization and normalization. The classification problem was formulated as a multi-label binary classification model, which used convolutional neural network architecture to make a decision on whether an abnormality was present or not in the chest radiographs. The classification model was evaluated using specificity, sensitivity, and Area Under Curve (AUC) score as the parameter. A prototype of the classification model was implemented using Keras Open source deep learning framework in Python Programming Language. The AUC ROC curve of the model was able to classify Atelestasis, Support devices, Pleural effusion, Pneumonia, A normal CXR (no finding), Pneumothorax, and Consolidation. However, Lung opacity and Cardiomegaly had a probability of less than 0.5 and thus were classified as absent. Precision, recall, and F1 score values were 0.78; this implies that the number of False Positive and False Negative is the same, revealing some measure of label imbalance in the dataset. The study concluded that the developed model is sufficient to classify abnormalities present in chest radiographs into present or absent.

Keywords: transfer learning, convolutional neural network, radiograph, classification, multi-label

Procedia PDF Downloads 86

Authors: Akshada Amit Koparde, Candrakant S. Magdum

Abstract:

Objective: To investigate the anti-arthritic activity of chloroform extract and Isolate 1 of Eulophia ochreata Lindl and dichloromethane extract and Isolate 2 of Zingiber cassumunar Roxb in adjuvant arthritic (AA) rat model induced by Freund’s complete adjuvant (FCA). Methods: Forty two healthy albino rats were selected and randomly divided into six groups. Freund’s complete adjuvant (FCA) was used to induce arthritis and then treated with chloroform extract, isolate 1 and dichloromethane extract, isolate 2 for 28 days. The various parameters like paw volume, haematological parameters (RBC, WBC, Hb and ESR), were studied. Structural elucidation of active constituents isolate 1 and isolate 2 from Eulophia ochreata Lindl and Zingiber cassumunar Roxb will be done using GCMS and H1NMR. Results: In FCA induced arthritic rats, there was significant increase in rat paw volume whereas chloroform extract and Isolate 1 of Eulophia ochreata Lindl and dichloromethane extract and Isolate 2 of Zingiber cassumunar Roxb treated groups showed strong significant reduction in paw volume. The altered haematological parameters in the arthritic rats were significantly recovered to near normal by the treatment with extracts at the dose of 200 mg/kg. Further histopathological studies revealed the anti-arthritic activity of Eulophia ochreata Lindl and Zingiber cassumunar Roxb by preventing cartilage and bone destruction of the arthritic joints of AA rats. Conclusion: Extracts and isolates of Eulophia ochreata Lindl and Zingiber cassumunar Roxb have shown anti-arthritic activity. Decrease in paw volume and normalization of haematological abnormalities in adjuvant induced arthritic rats is significantly seen in the experiment. Further histopathological studies confirmed the anti-arthritic activity of Eulophia ochreata Lindl and Zingiber cassumunar Roxb.

Keywords: arthritis, Eulophia ochreata Lindl, Freund's complete adjuvant, paw volume, Zingiber cassumunar Roxb

Procedia PDF Downloads 143

Authors: J. Wicks

Abstract:

Aim: To develop a simple methodology to allow collected heart rate (HR) data from inexpensive wearable devices to be expressed in a suitable format (METs) to quantitate occupational (and recreational) activity. Introduction: Assessment of occupational activity is commonly done by utilizing questionnaires in combination with prescribed MET levels of a vast range of previously measured activities. However for any individual the intensity of performing a specific activity can vary significantly. Ideally objective measurement of individual activity is preferred. Though there are a wide range of HR recording devices there is a distinct lack methodology to allow processing of collected data to quantitate energy expenditure (EE). The HR index equation expresses METs in relation to relative HR i.e. the ratio of activity HR to resting HR. The use of this equation provides a simple utility for objective measurement of EE. Methods: During a typical occupational work period of approximately 8 hours HR data was recorded using a Polar RS 400 wrist monitor. Recorded data was downloaded to a Windows PC and non HR data was stripped from the ASCII file using ‘Notepad’. The HR data was exported to a spread sheet program and sorted by HR range into a histogram format. Three HRs were determined, namely a resting HR (the HR delimiting the lowest 30 minutes of recorded data), a mean HR and a peak HR (the HR delimiting the highest 30 minutes of recorded data). HR indices were calculated (mean index equals mean HR/rest HR and peak index equals peak HR/rest HR) with mean and peak indices being converted to METs using the HR index equation. Conclusion: Inexpensive HR recording devices can be utilized to make reasonable estimates of occupational (or recreational) EE suitable for large scale demographic screening by utilizing the HR index equation. The intrinsic value of the HR index equation is that it is independent of factors that influence absolute HR, namely fitness, smoking and beta-blockade.

Keywords: energy expenditure, heart rate histograms, heart rate index, occupational activity

Procedia PDF Downloads 272

Authors: Asawari Alok Pawaskar

Abstract:

This study was aimed at investigating quality assurance (QA) done with IBA matrix, the discrepan­cies observed for helical tomotherapy plans. A selection of tomotherapy plans that initially failed the with Matrix process was chosen for this investigation. These plans failed the fluence analysis as assessed using gamma criteria (3%, 3 mm). Each of these plans was modified (keeping the planning constraints the same), beamlets rebatched and reoptimized. By increasing and decreasing the modula­tion factor, the fluence in a circumferential plane as measured with a diode array was assessed. A subset of these plans was investigated using varied pitch values. Factors for each plan that were examined were point doses, fluences, leaf opening times, planned leaf sinograms, and uniformity indices. In order to ensure that the treatment constraints remained the same, the dose-volume histograms (DVHs) of all the modulated plans were compared to the original plan. It was observed that a large increase in the modulation factor did not significantly improve DVH unifor­mity, but reduced the gamma analysis pass rate. This also increased the treatment delivery time by slowing down the gantry rotation speed which then increases the maximum to mean non-zero leaf open time ratio. Increasing and decreasing the pitch value did not substantially change treatment time, but the delivery accuracy was adversely affected. This may be due to many other factors, such as the complexity of the treatment plan and site. Patient sites included in this study were head and neck, breast, abdomen. The impact of leaf tim­ing inaccuracies on plans was greater with higher modulation factors. Point-dose measurements were seen to be less susceptible to changes in pitch and modulation factors. The initial modulation factor used by the optimizer, such that the TPS generated ‘actual’ modulation factor within the range of 1.4 to 2.5, resulted in an improved deliverable plan.

Keywords: dose volume histogram, modulation factor, IBA matrix, tomotherapy

Procedia PDF Downloads 150

Authors: Souvik Phadikar, Nidul Sinha, Rajdeep Ghosh

Abstract:

In brainwaves research using electroencephalogram (EEG) signals, finding the most relevant and effective feature set for identification of activities in the human brain is a big challenge till today because of the random nature of the signals. The feature extraction method is a key issue to solve this problem. Finding those features that prove to give distinctive pictures for different activities and similar for the same activities is very difficult, especially for the number of activities. The performance of a classifier accuracy depends on this quality of feature set. Further, more number of features result in high computational complexity and less number of features compromise with the lower performance. In this paper, a novel idea of the selection of optimal feature set using a heuristically optimized deep autoencoder is presented. Using various feature extraction methods, a vast number of features are extracted from the EEG signals and fed to the autoencoder deep neural network. The autoencoder encodes the input features into a small set of codes. To avoid the gradient vanish problem and normalization of the dataset, a meta-heuristic search algorithm is used to minimize the mean square error (MSE) between encoder input and decoder output. To reduce the feature set into a smaller one, 4 hidden layers are considered in the autoencoder network; hence it is called Heuristically Optimized Deep Autoencoder (HO-DAE). In this method, no features are rejected; all the features are combined into the response of responses of the hidden layer. The results reveal that higher accuracy can be achieved using optimal reduced features. The proposed HO-DAE is also compared with the regular autoencoder to test the performance of both. The performance of the proposed method is validated and compared with the other two methods recently reported in the literature, which reveals that the proposed method is far better than the other two methods in terms of classification accuracy.

Keywords: autoencoder, brainwave signal analysis, electroencephalogram, feature extraction, feature selection, optimization

Procedia PDF Downloads 94

Authors: Tomas Premoli, Sareh Rowlands

Abstract:

In this study, we investigate the impact of various convolutional neural network (CNN) architectures on the accuracy of diagnosing Alzheimer’s disease (AD) using patient MRI scans. Alzheimer’s disease is a debilitating neurodegenerative disorder that affects millions worldwide. Early, accurate, and non-invasive diagnostic methods are required for providing optimal care and symptom management. Deep learning techniques, particularly CNNs, have shown great promise in enhancing this diagnostic process. We aim to contribute to the ongoing research in this field by comparing the effectiveness of different CNN architectures and providing insights for future studies. Our methodology involved preprocessing MRI data, implementing multiple CNN architectures, and evaluating the performance of each model. We employed intensity normalization, linear registration, and skull stripping for our preprocessing. The selected architectures included VGG, ResNet, and DenseNet models, all implemented using the Keras library. We employed transfer learning and trained models from scratch to compare their effectiveness. Our findings demonstrated significant differences in performance among the tested architectures, with DenseNet201 achieving the highest accuracy of 86.4%. Transfer learning proved to be helpful in improving model performance. We also identified potential areas for future research, such as experimenting with other architectures, optimizing hyperparameters, and employing fine-tuning strategies. By providing a comprehensive analysis of the selected CNN architectures, we offer a solid foundation for future research in Alzheimer’s disease diagnosis using deep learning techniques. Our study highlights the potential of CNNs as a valuable diagnostic tool and emphasizes the importance of ongoing research to develop more accurate and effective models.

Keywords: Alzheimer’s disease, convolutional neural networks, deep learning, medical imaging, MRI

Procedia PDF Downloads 45

Authors: B. Yermekbayeva, A. Gulyayaev, T. Nurgozhin, C. Bektur

Abstract:

Cytisine aminophosphonate under the name "Cytafat" was approved for clinical trials in Republic of Kazakhstan as a putative liver protecting drug for the treatment of acute toxic hepatitis. A method of conducting the clinical trial is a double blind study. Total number of patients -71, aged from 16 to 56 years. Research on healthy volunteers determined the maximal tolerable doze of "Cytafat" as 200 mg/kg. Side effects when administered at high dozes (100-200 mg/kg) are tachycardia and increase of arterial blood pressure. The drug is tested in the treatment of 28 patients with a syndrome of hepatocellular failure (a poisoning with substitutes of alcohol, rat poison, or medical products). "Cytafat" was intravenously administered at a dose of 10 mg/kg in 200 ml of 5 % glucose solution once daily. The number of administrations: 1-3. In the comparison group, 23 patients were treated intravenously once a day with “Essenciale H” at a dose of 10 ml. 20 patients received a placebo (10 ml of glucose intravenously). In all cases of toxic hepatopathology the significant positive clinical effect of the testing drug distinguishable from placebo and surpassing the alternative was observed. Within a day after administration a sharp reduction of cytolitic syndrome parameters (ALT, AST, alkaline phosphatase, thymol turbidity test, GGT) was registered, a reduction of the severity of cholestatic syndrome (bilirubin decreased) was recorded, significantly decreased indices of lipid peroxidation. The following day, in all cases the positive dynamics was determined with ultrasound study (reduction of diffuse changes and events of reactive pancreatitis), hepatomegaly disappeared. Normalization of all parameters occurred in 2-3 times faster, than when using the drug "Essenciale H" and placebo. Average term of elimination of toxic hepatopathy when using the drug "Cytafat" -2,8 days, "Essenciale H" -7,2 days, and placebo -10,6 days. The new drug "Cytafat" has expressed cytoprotective properties.

Keywords: cytisine, cytoprotection, hepatopathy, hepatoprotection

Procedia PDF Downloads 339

Authors: Muneeb Ullah, Daishihan, Xiadong Young

Abstract:

Deep learning plays a crucial role in identifying COVID-19 and preventing its spread. To improve the accuracy of COVID-19 diagnoses, it is important to have access to a sufficient number of training images of CXRs (chest X-rays) depicting the disease. However, there is currently a shortage of such images. To address this issue, this paper introduces COVID-19 GAN, a model that uses generative adversarial networks (GANs) to generate realistic CXR images of COVID-19, which can be used to train identification models. Initially, a generator model is created that uses digressive channels to generate images of CXR scans for COVID-19. To differentiate between real and fake disease images, an efficient discriminator is developed by combining the dense connectivity strategy and instance normalization. This approach makes use of their feature extraction capabilities on CXR hazy areas. Lastly, the deep regret gradient penalty technique is utilized to ensure stable training of the model. With the use of 4,062 grape leaf disease images, the Leaf GAN model successfully produces 8,124 COVID-19 CXR images. The COVID-19 GAN model produces COVID-19 CXR images that outperform DCGAN and WGAN in terms of the Fréchet inception distance. Experimental findings suggest that the COVID-19 GAN-generated CXR images possess noticeable haziness, offering a promising approach to address the limited training data available for COVID-19 model training. When the dataset was expanded, CNN-based classification models outperformed other models, yielding higher accuracy rates than those of the initial dataset and other augmentation techniques. Among these models, ImagNet exhibited the best recognition accuracy of 99.70% on the testing set. These findings suggest that the proposed augmentation method is a solution to address overfitting issues in disease identification and can enhance identification accuracy effectively.

Keywords: classification, deep learning, medical images, CXR, GAN.

Procedia PDF Downloads 54

Authors: Y. Smirlis, G. Koronakos, S. Plitsos

Abstract:

Shipbrokers assist the ship companies in chartering or selling and buying vessels, acting as intermediates between them and the market. They facilitate deals, providing their expertise, negotiating skills, and knowledge about ship market bargains. Their role is very important as it affects the profitability and market position of a shipping company. Due to their significant contribution, the shipping companies have to employ systematic procedures to evaluate the shipbrokers’ services in order to select the best and, consequently, to achieve the best deals. Towards this, in this paper, we consider shipbrokers as financial service providers, and we formulate the problem of evaluating and selecting shipbrokers’ services as a multi-criteria decision making (MCDM) procedure. The proposed methodology comprises a first normalization step to adjust different scales and orientations of the criteria and a second step that includes the mathematical model to evaluate the performance of the shipbrokers’ services involved in the assessment. The criteria along which the shipbrokers are assessed may refer to their size and reputation, the potential efficiency of the services, the terms and conditions imposed, the expenses (e.g., commission – brokerage), the expected time to accomplish a chartering or selling/buying task, etc. and according to our modelling approach these criteria may be assigned different importance. The mathematical programming model performs a comparative assessment and estimates for the shipbrokers involved in the evaluation, a relative score that ranks the shipbrokers in terms of their potential performance. To illustrate the proposed methodology, we present a case study in which a shipping company evaluates and selects the most suitable among a number of sale and purchase (S&P) brokers. Acknowledgment: This study is supported by the OptiShip project, implemented within the framework of the National Recovery Plan and Resilience “Greece 2.0” and funded by the European Union – NextGenerationEU programme.

Keywords: shipbrokers, multi-criteria decision making, mathematical programming, service-provider selection

Procedia PDF Downloads 52

Authors: Mikiyas Adugna

Abstract:

Effort estimation is among the activities carried out in software development processes. An accurate model of estimation leads to project success. The method of agile effort estimation is a complex task because of the dynamic nature of software development. Researchers are still conducting studies on agile effort estimation to enhance prediction accuracy. Due to these reasons, we investigated and proposed a model on LASSO and Elastic Net regression to enhance estimation accuracy. The proposed model has major components: preprocessing, train-test split, training with default parameters, and cross-validation. During the preprocessing phase, the entire dataset is normalized. After normalization, a train-test split is performed on the dataset, setting training at 80% and testing set to 20%. We chose two different phases for training the two algorithms (Elastic Net and LASSO) regression following the train-test-split. In the first phase, the two algorithms are trained using their default parameters and evaluated on the testing data. In the second phase, the grid search technique (the grid is used to search for tuning and select optimum parameters) and 5-fold cross-validation to get the final trained model. Finally, the final trained model is evaluated using the testing set. The experimental work is applied to the agile story point dataset of 21 software projects collected from six firms. The results show that both Elastic Net and LASSO regression outperformed the compared ones. Compared to the proposed algorithms, LASSO regression achieved better predictive performance and has acquired PRED (8%) and PRED (25%) results of 100.0, MMRE of 0.0491, MMER of 0.0551, MdMRE of 0.0593, MdMER of 0.063, and MSE of 0.0007. The result implies LASSO regression algorithm trained model is the most acceptable, and higher estimation performance exists in the literature.

Keywords: agile software development, effort estimation, elastic net regression, LASSO

Procedia PDF Downloads 27

Authors: Soumaya Mrabet, Imen Akkari, Amira Atig, Elhem Ben Jazia

Abstract:

Introduction: Celiac disease (CD) is a chronic autoimmune inflammatory enteropathy caused by gluten. The clinical presentation is very variable. Malabsorption in the MC is responsible for an alteration of the bone metabolism. Our purpose is to study the osteoarticular manifestations related to this condition. Material and methods: It is a retrospective study of 41 cases of CD diagnosed on clinical, immunological, endoscopic and histological arguments, in the Internal Medicine and Gastroenterology Department of Farhat Hached Hospital between September 2005 and January 2016. Results: Osteoarticular manifestations were found in 9 patients (22%) among 41 patients presenting CD. These were 7 women and 2 men with an average age of 35.7 years (25 to 67 years). These manifestations were revelatory of CD in 3 cases. Abdominal pain and diarrhea were present in 6 cases. Inflammatory polyarthralgia of wrists and knees has been reported in 7 patients. Mechanical mono arthralgia was noted in 2 patients. Biological tests revealed microcytic anemia by iron deficiency in 7 cases, hypocalcemia in 5 cases, Hypophosphatemia in 3 cases and elevated alkaline phosphatases in 3 cases. Upper gastrointestinal endoscopy with duodenal biopsy found villous atrophy in all cases. In immunology, Anti-transglutaminase antibodies were positive in all patients, Anti-endomysium in 7 cases. Measurement of bone mineral density (BMD) by biphotonic X-ray absorptiometer with evaluation of the T-score and the Z-score was performed in Twenty patients (48.8%). It was normal in 7 cases (33%) and showed osteopenia in 5 patients (25%) and osteoporosis in 2 patients (10%). All patients were treated with a Gluten-free diet associated with vitamin D and calcium substitution in 5 cases. The evolution was favorable in all cases with reduction of bone pain and normalization of the phosphocalcic balance. Conclusion: The bone impact of CD is frequent but often asymptomatic. Patients with CD should be evaluated by the measurement of bone mineral density and monitored for calcium and vitamin D deficiencies.

Keywords: bone mineral density, celiac disease, osteoarticular manifestations, vitamin D and calcium

Procedia PDF Downloads 299

Authors: N. C. Ghosh, Niladri Das, Prolay Mondal, Ranajit Ghosh

Abstract:

Groundwater quality deterioration due to human activities has become a prime factor of modern life. The major concern of the study is to access spatial variation of groundwater quality and to identify the sources of groundwater chemicals and its impact on human health of the concerned area. Multivariate statistical techniques, cluster, principal component analysis, and hydrochemical fancies are been applied to measure groundwater quality data on 14 parameters from 107 sites distributed randomly throughout the Birbhum district. Five factors have been extracted using Varimax rotation with Kaiser Normalization. The first factor explains 27.61% of the total variance where high positive loading have been concentrated in TH, Ca, Mg, Cl and F (Fluoride). In the studied region, due to the presence of basaltic Rajmahal trap fluoride contamination is highly concentrated and that has an adverse impact on human health such as fluorosis. The second factor explains 24.41% of the total variance which includes Na, HCO₃, EC, and SO₄. The last factor or the fifth factor explains 8.85% of the total variance, and it includes pH which maintains the acidic and alkaline character of the groundwater. Hierarchical cluster analysis (HCA) grouped the 107 sampling station into two clusters. One cluster having high pollution and another cluster having less pollution. Moreover hydromorphological facies viz. Wilcox diagram, Doneen’s chart, and USSL diagram reveal the quality of the groundwater like the suitability of the groundwater for irrigation or water used for drinking purpose like permeability index of the groundwater, quality assessment of groundwater for irrigation. Gibb’s diagram depicts that the major portion of the groundwater of this region is rock dominated origin, as the western part of the region characterized by the Jharkhand plateau fringe comprises basalt, gneiss, granite rocks.

Keywords: correlation, factor analysis, hydrological facies, hydrochemistry

Procedia PDF Downloads 184

Authors: Ana Paola Echavarria, Mariuxi Medina, Haydelba D'Armas, Carmita Jaramillo, Diana San Martin

Abstract:

The use of medicinal plants is one of the oldest and most extended medical therapies that goes back to prehistoric times, and nowadays, they are also used in the preparation of phytopharmaceuticals with options to cure diseases. The test for the determination of fungi was carried out in the Pharmacy Pilot Plant (treatment of the leaves of the plant species) and the Microbiology Laboratory (determination of fungi of the plant species, using growth medium called Sabouraud agar plus the vegetal sample), of the Academic Unit of Chemical Sciences and Health, of the Universidad Tecnica de Machala. Subsequently, colony counting was performed, both macroscopic, which is determined in the growth medium of the seeding, and microscopic, to identify the germinative forms using blue lactophenol. The procedure was repeated in duplicate to replicate the results data. The determination of the total fungal content of the following plant species was evaluated: Cymbopogon citratus (lemon verbena), Melissa officinalis (lemon balm), Taraxacum officinale (dandelion), Artemisia absinthium (absinthe), Piper carpunya (guaviduca), Moringa oleifera (moringa), Coriandrum sativum (coriander), Momordica charantia (achochilla), Borago officinalis (borage), Aloysia citriodora (cedron), Ambrosia artemisifolia (altamisa) and Ageratum conyzoides (mastrante). The results obtained showed that all the samples of the twelve plant species studied developed filamentous fungi, with great variability of them, within the permissible limits and contemplated by the Ecuadorian Institute of Normalization (INEN), being suitable as raw material for its use in the preparation of nutraceuticals and medicinal products or phytodrugs; with the exception of A. conyzoides (mastranto) which is the only species that exceeds the regulation in the average of dilutions.

Keywords: colonies, fungi, medicinal plants, microbiological quality, Sabouraud agar

Procedia PDF Downloads 118

Authors: Nur Aziza Luxfiati

Abstract:

Clustering is partitioning data sets into sub-sets or groups in such a way that elements certain properties have shared property settings with a high level of similarity within one group and a low level of similarity between groups. . The K-Means algorithm is one of thealgorithmsclustering as a grouping tool that is most widely used in scientific and industrial applications because the basic idea of the kalgorithm is-means very simple. In this research, applying the technique of clustering using the k-means algorithm as a method of solving the problem of national development imbalances between regions in Central Java Province based on macroeconomic indicators. The data sample used is secondary data obtained from the Central Java Provincial Statistics Agency regarding macroeconomic indicator data which is part of the publication of the 2019 National Socio-Economic Survey (Susenas) data. score and determine the number of clusters (k) using the elbow method. After the clustering process is carried out, the validation is tested using themethodsBetween-Class Variation (BCV) and Within-Class Variation (WCV). The results showed that detection outlier using z-score normalization showed no outliers. In addition, the results of the clustering test obtained a ratio value that was not high, namely 0.011%. There are two district/city clusters in Central Java Province which have economic similarities based on the variables used, namely the first cluster with a high economic level consisting of 13 districts/cities and theclustersecondwith a low economic level consisting of 22 districts/cities. And in the cluster second, namely, between low economies, the authors grouped districts/cities based on similarities to macroeconomic indicators such as 20 districts of Gross Regional Domestic Product, with a Poverty Depth Index of 19 districts, with 5 districts in Human Development, and as many as Open Unemployment Rate. 10 districts.

Keywords: clustering, K-Means algorithm, macroeconomic indicators, inequality, national development

Procedia PDF Downloads 131

Authors: Alejandro Arnaldo Barroso Martinez

Abstract:

Unlike sex, which is constituted by anatomic-physiological differences, gender is a social construction. Our thoughts and behaviors as females and males are not etched in stone by our biology but rather from how society expects us to think and behave based on our sex assignment in the womb. Social expectations, values, and roles are taken on by individuals and shape the ways considered acceptable and linked to our bodies, feelings, and interpersonal relationships. Furthermore, these evolve into dire consequences for those who do not meet these disciplinary, economic, and cultural standards. Then, the social learning of gender identity implies the individual’s psychological sense of being, and it might be highly linked to a sense of life and suicide attempts. As a result, suicide has been considered a gender issue with differences in the rates and means used by men and women worldwide. Nevertheless, there has been a misunderstanding of the meaning of being male or female in a particular context and how it becomes a risk process for suicide attempts. For this reason, the general objective of the current research is to explain how this process occurs in Cuba. From a Critical Sociology and Social Psychology, a qualitative methodology was developed through six case studies and qualitative in-depth interviews. The analysis is focused on the sequence and interplay between two dimensions of meaning: signifiers and voices. Findings show that the risk process of suicide attempts in Cuba means some patriarchal beliefs and practices as part of informal educational models and some positivist practices in mental health attention. Findings also show that community relations create a sense of belonging, and it is a protection against suicide attempts in Cuba. Those frames of signifiers and voices explain in both males and females but differently when and how they are suffering from isolation, violence, the normalization of emotional awareness, and emotional distress expression. Suicide prevention programs should take gender learning into account as a cultural process.

Keywords: social constructions, gender identity, meanings, suicide attempt

Procedia PDF Downloads 186

Authors: Wade Ghribi, Abdelmoty M. Ahmed, Ahmed Said Badawy, Belgacem Bouallegue

Abstract:

In higher education institutions, the most pressing priority is to improve student performance and retention. Large volumes of student data are used in Educational Data Mining techniques to find new hidden information from students' learning behavior, particularly to uncover the early symptom of at-risk pupils. On the other hand, data with noise, outliers, and irrelevant information may provide incorrect conclusions. By identifying features of students' data that have the potential to improve performance prediction results, comparing and identifying the most appropriate ensemble learning technique after preprocessing the data, and optimizing the hyperparameters, this paper aims to develop a reliable students' performance prediction model for Higher Education Institutions. Data was gathered from two different systems: a student information system and an e-learning system for undergraduate students in the College of Computer Science of a Saudi Arabian State University. The cases of 4413 students were used in this article. The process includes data collection, data integration, data preprocessing (such as cleaning, normalization, and transformation), feature selection, pattern extraction, and, finally, model optimization and assessment. Random Forest, Bagging, Stacking, Majority Vote, and two types of Boosting techniques, AdaBoost and XGBoost, are ensemble learning approaches, whereas Decision Tree, Support Vector Machine, and Artificial Neural Network are supervised learning techniques. Hyperparameters for ensemble learning systems will be fine-tuned to provide enhanced performance and optimal output. The findings imply that combining features of students' behavior from e-learning and students' information systems using Majority Vote produced better outcomes than the other ensemble techniques.

Keywords: educational data mining, student performance prediction, e-learning, classification, ensemble learning, higher education

Procedia PDF Downloads 77

Authors: Hediyeh Talebi, Shokoofeh Ghiam, Changiz Eslahchi

Abstract:

Traditionally, Alzheimer’s disease research has focused on genes with significant fold changes, potentially neglecting subtle but biologically important alterations. Our study introduces an integrative approach that highlights genes crucial to underlying biological processes, regardless of their fold change magnitude. Alzheimer's Single-cell RNA-seq data related to the peripheral blood mononuclear cells (PBMC) was extracted from the Gene Expression Omnibus (GEO). After quality control, normalization, scaling, batch effect correction, and clustering, differentially expressed genes (DEGs) were identified with adjusted p-values less than 0.05. These DEGs were categorized based on cell-type, resulting in four datasets, each corresponding to a distinct cell type. To distinguish between cells from healthy individuals and those with Alzheimer's, an adversarial autoencoder with a classifier was employed. This allowed for the separation of healthy and diseased samples. To identify the most influential genes in this classification, the weight matrices in the network, which includes the encoder and classifier components, were multiplied, and focused on the top 20 genes. The analysis revealed that while some of these genes exhibit a high fold change, others do not. These genes, which may be overlooked by previous methods due to their low fold change, were shown to be significant in our study. The findings highlight the critical role of genes with subtle alterations in diagnosing Alzheimer's disease, a facet frequently overlooked by conventional methods. These genes demonstrate remarkable discriminatory power, underscoring the need to integrate biological relevance with statistical measures in gene prioritization. This integrative approach enhances our understanding of the molecular mechanisms in Alzheimer’s disease and provides a promising direction for identifying potential therapeutic targets.

Keywords: alzheimer's disease, single-cell RNA-seq, neural networks, blood biomarkers

Procedia PDF Downloads 33