Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689901
Umme Afruz, Md. Ahasan Kabir
Wireless communication is a revolutionary technology that the world is grappling with right now. Because of the increased use of wireless networks and electrical apparatus, wireless body area networks are becoming more widespread. By placing multiple devices on the human body, WBAN has established a connection between them. The wearable antenna is utilized to improve various WBAN applications. This study shows a low-profile wearable antenna through WBAN for ongoing monitoring of critical human indicators like blood pressure, pulse rate, as well as skin temperature. A small, low-profile, and flexible antenna made of FR-4 material is presented. The antenna works with an operating frequency of 2.45 GHz and has a return loss of less than -10dB. The proposed antenna’s overall dimensions are 40x38mm2. The designed antenna’s gain, directivity, VSWR, bandwidth, and SAR are all simulated utilizing CST software. The antenna has a gain of 1.85 dB, a return loss of -20.18dB, a bandwidth of 594 MHz based on $|S_{11}|leq-10dB$, a directivity of 2.3 dB, and radiation efficiency of 90.2 percent, according to the simulation results. This antenna can become a trustworthy selection for WBAN applications in the ISM band due to its satisfactory performance.
{"title":"Design of Wearable Microstrip Patch Antenna for Wireless Body Area Network","authors":"Umme Afruz, Md. Ahasan Kabir","doi":"10.1109/ICCIT54785.2021.9689901","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689901","url":null,"abstract":"Wireless communication is a revolutionary technology that the world is grappling with right now. Because of the increased use of wireless networks and electrical apparatus, wireless body area networks are becoming more widespread. By placing multiple devices on the human body, WBAN has established a connection between them. The wearable antenna is utilized to improve various WBAN applications. This study shows a low-profile wearable antenna through WBAN for ongoing monitoring of critical human indicators like blood pressure, pulse rate, as well as skin temperature. A small, low-profile, and flexible antenna made of FR-4 material is presented. The antenna works with an operating frequency of 2.45 GHz and has a return loss of less than -10dB. The proposed antenna’s overall dimensions are 40x38mm2. The designed antenna’s gain, directivity, VSWR, bandwidth, and SAR are all simulated utilizing CST software. The antenna has a gain of 1.85 dB, a return loss of -20.18dB, a bandwidth of 594 MHz based on $|S_{11}|leq-10dB$, a directivity of 2.3 dB, and radiation efficiency of 90.2 percent, according to the simulation results. This antenna can become a trustworthy selection for WBAN applications in the ISM band due to its satisfactory performance.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125310344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689828
Muhtasim Shafi Kader, Fizar Ahmed, Jobeda Akter
Nowadays poultry is the best production of animal protein. With the amazing food diversity of Bangladesh, poultry chicken has a great impact on our daily life. But some major diseases are hampering this industry frequently. Serpentine illness such as infected bursal disease is more prevalent followed by colibacillosis, Newcastle disease, salmonellosis, chronic breathing disease, Avian Influenza, coccidiosis, aspergillosis, omphalitis, fowl pox, nutritional deficiency. Machine learning can be a useful health care way and also poultry disease precaution and detection. In advanced computer science diseases like Avian Influenza, Newcastle Disease are harmful to chicken. In order to prevent harmful consequences, it is important to concentrate about poultry infection on our very initial stage. We use a few qualities to evaluate our analysis regarding poultry illness and this attribute is one of the key items of the following disease. Perhaps we implement eleven machine classifiers to measure analysis by employing the following technologies, Logistic Regression Classifier, Naive Bayes Classifier, Multilayer Classifier, Stochastic Gradient Classifier, r Random Forest classifier, Bagging Classifier, Decision Tree Classifier, K Nearest Neighbor Classifier, XGB Classifier, AdaBoost Classifier & Gradient Boosting Classifier. The method we employed here gives maximum precision. Decision Tree Classifier has the best outcome yet.
{"title":"Machine Learning Techniques to Precaution of Emerging Disease in the Poultry Industry","authors":"Muhtasim Shafi Kader, Fizar Ahmed, Jobeda Akter","doi":"10.1109/ICCIT54785.2021.9689828","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689828","url":null,"abstract":"Nowadays poultry is the best production of animal protein. With the amazing food diversity of Bangladesh, poultry chicken has a great impact on our daily life. But some major diseases are hampering this industry frequently. Serpentine illness such as infected bursal disease is more prevalent followed by colibacillosis, Newcastle disease, salmonellosis, chronic breathing disease, Avian Influenza, coccidiosis, aspergillosis, omphalitis, fowl pox, nutritional deficiency. Machine learning can be a useful health care way and also poultry disease precaution and detection. In advanced computer science diseases like Avian Influenza, Newcastle Disease are harmful to chicken. In order to prevent harmful consequences, it is important to concentrate about poultry infection on our very initial stage. We use a few qualities to evaluate our analysis regarding poultry illness and this attribute is one of the key items of the following disease. Perhaps we implement eleven machine classifiers to measure analysis by employing the following technologies, Logistic Regression Classifier, Naive Bayes Classifier, Multilayer Classifier, Stochastic Gradient Classifier, r Random Forest classifier, Bagging Classifier, Decision Tree Classifier, K Nearest Neighbor Classifier, XGB Classifier, AdaBoost Classifier & Gradient Boosting Classifier. The method we employed here gives maximum precision. Decision Tree Classifier has the best outcome yet.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114618753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689784
Md. Nazmul Islam, M. Hasan, Abdul Kadar Muhammad Masum, Md. Zia Uddin, Md. Golam Rabiul Alam
Covid 19 continues to have a catastrpoic effect on the world, causing terrible spots to appear all over the place. Due to global epidemics and doctor and healthcare personel shortages, developing an AI-based system to detect COVID in a timely and cost-effective method has become a requirement. It is also essential to detect covid from chest X-ray and CT radiographs due to their accuracy in detecting lung infection and as well as to understand the severity. Moreover, though the number of infected people around the globe is enormous, the amount of covid data set to build an AI system is scarce and scattered. In this letter, we presented a Chest CT scan data (HRCT) set for Covid and healthy patients considering a varying range of severity of COVID, which we published on kaggle, that can assist other researchers to contribute to healthcare AI. We also developed three deep learning approaches for detecting covid quickly and cheaply. Our three transfer learning-based approaches, Inception v3, Resnet 50, and VGG16, achieve accuracy of 99.8%, 91.3%, and 99.3%, respectively on unseen data. We delve deeper into the black boxes of those models to demonstrate how our model comes to a certain conclusion, and we found that, despite the low accuracy of the model based on VGG16, it detects the covid spot of images well, which we believe may further assist doctors in visualizing which regions are affected.
{"title":"Demystify the Black-box of Deep Learning Models for COVID-19 Detection from Chest CT Radiographs","authors":"Md. Nazmul Islam, M. Hasan, Abdul Kadar Muhammad Masum, Md. Zia Uddin, Md. Golam Rabiul Alam","doi":"10.1109/ICCIT54785.2021.9689784","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689784","url":null,"abstract":"Covid 19 continues to have a catastrpoic effect on the world, causing terrible spots to appear all over the place. Due to global epidemics and doctor and healthcare personel shortages, developing an AI-based system to detect COVID in a timely and cost-effective method has become a requirement. It is also essential to detect covid from chest X-ray and CT radiographs due to their accuracy in detecting lung infection and as well as to understand the severity. Moreover, though the number of infected people around the globe is enormous, the amount of covid data set to build an AI system is scarce and scattered. In this letter, we presented a Chest CT scan data (HRCT) set for Covid and healthy patients considering a varying range of severity of COVID, which we published on kaggle, that can assist other researchers to contribute to healthcare AI. We also developed three deep learning approaches for detecting covid quickly and cheaply. Our three transfer learning-based approaches, Inception v3, Resnet 50, and VGG16, achieve accuracy of 99.8%, 91.3%, and 99.3%, respectively on unseen data. We delve deeper into the black boxes of those models to demonstrate how our model comes to a certain conclusion, and we found that, despite the low accuracy of the model based on VGG16, it detects the covid spot of images well, which we believe may further assist doctors in visualizing which regions are affected.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114567664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689871
Hasan Bin Firoz, Nawshin Mannan Proma
The scientific community has been extensively studying different robotic aerial systems over the past few decades. Among them, vertical take-off and landing vehicles (VTOLs) such as Quadrotors have secured a special place. In many of their applications, a quadrotor needs to fly in an unknown environment without any human intervention. In order to guarantee the safety and efficiency of an autonomous flight, quadrotors need to track a pre-defined trajectory precisely. The ultimate goal of this research work is to design a deep neural network-based controller that can replace the classical PID controller with a view to achieving improved trajectory tracking performance. In the end, a comparison between the conventional controller and the proposed DNN based controller is presented to highlight the improvement in terms of trajectory tracking performance.
{"title":"Deep Neural Network Based Controller Design for Improved Trajectory Tracking of Quadrotor Unmanned Aerial Vehicles","authors":"Hasan Bin Firoz, Nawshin Mannan Proma","doi":"10.1109/ICCIT54785.2021.9689871","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689871","url":null,"abstract":"The scientific community has been extensively studying different robotic aerial systems over the past few decades. Among them, vertical take-off and landing vehicles (VTOLs) such as Quadrotors have secured a special place. In many of their applications, a quadrotor needs to fly in an unknown environment without any human intervention. In order to guarantee the safety and efficiency of an autonomous flight, quadrotors need to track a pre-defined trajectory precisely. The ultimate goal of this research work is to design a deep neural network-based controller that can replace the classical PID controller with a view to achieving improved trajectory tracking performance. In the end, a comparison between the conventional controller and the proposed DNN based controller is presented to highlight the improvement in terms of trajectory tracking performance.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"28 30","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120854837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689876
Md. Nazmul Islam, Mahmudul Hasan
In recent era, people are getting more attracted to micro blogs and social media to share their daily activities and express feelings and opinions. Machine learning based sentiment analysis becomes immensely popular to judge the feelings about a particular content on how positive or negative their feelings and opinions are before taking important decisions. In this paper, we propose an effective and combined machine learning approach with an enhanced hybrid feature vector space of latent concepts and external information features. The latent concepts are prepared by a supervised machine learning approach, and the external information features, estimating the quality of the information shared in the documents, are classified by the unsupervised rule-based learning approach. A Random Forest ensemble method has been utilized to build a classifier model, and some standard performance measures such as accuracy, precision, recall, f1-score and Cohen’s Kappa value have been taken into account to analyze the performance. The novelty of this paper lies in the hybridization of feature vector space of latent concepts and external information features along with the Random Forest ensemble classifier. Based on the analyses, the proposed approach outperforms its counterparts as well as provides better outcomes against other solo latent concept-oriented approaches.
{"title":"Hybrid Feature Vector Space based Ensemble Machine Learning Approach for Sentiment Analysis on Amazon Product Reviews","authors":"Md. Nazmul Islam, Mahmudul Hasan","doi":"10.1109/ICCIT54785.2021.9689876","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689876","url":null,"abstract":"In recent era, people are getting more attracted to micro blogs and social media to share their daily activities and express feelings and opinions. Machine learning based sentiment analysis becomes immensely popular to judge the feelings about a particular content on how positive or negative their feelings and opinions are before taking important decisions. In this paper, we propose an effective and combined machine learning approach with an enhanced hybrid feature vector space of latent concepts and external information features. The latent concepts are prepared by a supervised machine learning approach, and the external information features, estimating the quality of the information shared in the documents, are classified by the unsupervised rule-based learning approach. A Random Forest ensemble method has been utilized to build a classifier model, and some standard performance measures such as accuracy, precision, recall, f1-score and Cohen’s Kappa value have been taken into account to analyze the performance. The novelty of this paper lies in the hybridization of feature vector space of latent concepts and external information features along with the Random Forest ensemble classifier. Based on the analyses, the proposed approach outperforms its counterparts as well as provides better outcomes against other solo latent concept-oriented approaches.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126122557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689914
Md. Awlad Hossen Rony, K. Fatema, Md. Zahid Hasan
Plant disease classification is often accomplished by visual assessment or during research facility assessment which creates setbacks bringing about yield in loss when diagnosis is completed. Plant disease detection through an automated approach is advantageous because it minimizes the amount of monitoring required in large crop farms and identifies disease signs at an early stage, i.e., when they develop on plant leaves. Our suggested method adds to the automatic recognition of plant diseases through a series of processes that include pre-processing, analysis, and classification. In this study, an unsharp masking filter utilizes to process the blurred and the unsharpened part of the real images presents as a mask for producing a sharpened resulting image. As an image enhancement, a green fire blue filter is used to enrich the quality of images by increasing the contrast, removal the colors, and thresholding the images. For the verification of image quality, several statistics formulas such as PSNR, MSE, SSIM and SNR are calculated in the dataset. And finally, a proposed bottlenet18 deep learning architecture has been applied to classify three different Bottle gourd diseases as Anthracnose, Cercospora leaf spot, and Powdery mildew. In this work, we have measured the performance based on the performance matrices with variations of different optimizers and learning rates. The highest accuracy achieved by using the proposed BottleNet18 architecture is 93.9987% with Adam optimizer and 0.001 learning rate.
{"title":"BottleNet18: Deep Learning-Based Bottle Gourd Leaf Disease Classification","authors":"Md. Awlad Hossen Rony, K. Fatema, Md. Zahid Hasan","doi":"10.1109/ICCIT54785.2021.9689914","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689914","url":null,"abstract":"Plant disease classification is often accomplished by visual assessment or during research facility assessment which creates setbacks bringing about yield in loss when diagnosis is completed. Plant disease detection through an automated approach is advantageous because it minimizes the amount of monitoring required in large crop farms and identifies disease signs at an early stage, i.e., when they develop on plant leaves. Our suggested method adds to the automatic recognition of plant diseases through a series of processes that include pre-processing, analysis, and classification. In this study, an unsharp masking filter utilizes to process the blurred and the unsharpened part of the real images presents as a mask for producing a sharpened resulting image. As an image enhancement, a green fire blue filter is used to enrich the quality of images by increasing the contrast, removal the colors, and thresholding the images. For the verification of image quality, several statistics formulas such as PSNR, MSE, SSIM and SNR are calculated in the dataset. And finally, a proposed bottlenet18 deep learning architecture has been applied to classify three different Bottle gourd diseases as Anthracnose, Cercospora leaf spot, and Powdery mildew. In this work, we have measured the performance based on the performance matrices with variations of different optimizers and learning rates. The highest accuracy achieved by using the proposed BottleNet18 architecture is 93.9987% with Adam optimizer and 0.001 learning rate.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126340918","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689918
M. Hasan, Md. Ali Hossain, Azmain Yakin Srizon, Abu Sayeed, Mohiuddin Ahmed, Md Rakibul Haquek
Deep learning has remodeled the research aspect of facial recognition throughout the last decade by utilizing multiple processing layers to extract significant facial features. Although this emerging technology has achieved high performance for the face recognition problems, the dilemma of achieving low performance while training with a few samples per class has not been resolved yet. In this study, it has been shown that by utilizing retraining as a heuristic step, ResNet-50 based VGGFace architecture can enhance the performance of the face recognition scheme significantly. Multi-task Cascaded Convolutional Neural Networks have been utilized to crop faces first. The first training phase was completed by considering train samples from a combined dataset of 5-celebrity dataset, Georgia tech database, and three variants of KomNet datasets. The retraining of individual datasets further produced 94.41% test accuracy for the KomNet social media dataset and 100% test accuracy for the other four datasets.
{"title":"Improving Performance of a Pre-trained ResNet-50 Based VGGFace Recognition System by Utilizing Retraining as a Heuristic Step","authors":"M. Hasan, Md. Ali Hossain, Azmain Yakin Srizon, Abu Sayeed, Mohiuddin Ahmed, Md Rakibul Haquek","doi":"10.1109/ICCIT54785.2021.9689918","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689918","url":null,"abstract":"Deep learning has remodeled the research aspect of facial recognition throughout the last decade by utilizing multiple processing layers to extract significant facial features. Although this emerging technology has achieved high performance for the face recognition problems, the dilemma of achieving low performance while training with a few samples per class has not been resolved yet. In this study, it has been shown that by utilizing retraining as a heuristic step, ResNet-50 based VGGFace architecture can enhance the performance of the face recognition scheme significantly. Multi-task Cascaded Convolutional Neural Networks have been utilized to crop faces first. The first training phase was completed by considering train samples from a combined dataset of 5-celebrity dataset, Georgia tech database, and three variants of KomNet datasets. The retraining of individual datasets further produced 94.41% test accuracy for the KomNet social media dataset and 100% test accuracy for the other four datasets.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127239315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689892
Rubayea Ferdows, Fuad Ahmed, Md Rafiqul Islamy, A. Kamal
Social network (SN) applications such as Facebook, Twitter, Instagram, etc. provide many facilities that allow the user to connect, follow one another, share content, and influence them to engage in various activities in their personal lives. Sometimes it impacts their habits such as online buying, restaurant checkin, traveling, etc. Existing researchers have used a variety of approaches to identify these impacts on various topics, including fitness, psychological health, and so on. However, there is very few research that has been done for investigating individual expenditures. Thus, in this paper, we aim to 1) investigate the relationship between social media use and personal financial activities, 2) estimate personal expenditure based on various social media aspects such as checking into restaurants, buying clothes, traveling to new locations, doing something entertaining, and so on. We collected data through an online survey using social network platforms such as Facebook. We apply a causal model using propensity score-based inverse probability treatment weighting (IPTW) and a doubly robust estimator. We evaluate our approach by refuting the outcome. Finally, we find that social media usage has a significant impact on spending patterns.
{"title":"Exploring Causal Effect of Personal Financial Activities by Social Media Influences","authors":"Rubayea Ferdows, Fuad Ahmed, Md Rafiqul Islamy, A. Kamal","doi":"10.1109/ICCIT54785.2021.9689892","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689892","url":null,"abstract":"Social network (SN) applications such as Facebook, Twitter, Instagram, etc. provide many facilities that allow the user to connect, follow one another, share content, and influence them to engage in various activities in their personal lives. Sometimes it impacts their habits such as online buying, restaurant checkin, traveling, etc. Existing researchers have used a variety of approaches to identify these impacts on various topics, including fitness, psychological health, and so on. However, there is very few research that has been done for investigating individual expenditures. Thus, in this paper, we aim to 1) investigate the relationship between social media use and personal financial activities, 2) estimate personal expenditure based on various social media aspects such as checking into restaurants, buying clothes, traveling to new locations, doing something entertaining, and so on. We collected data through an online survey using social network platforms such as Facebook. We apply a causal model using propensity score-based inverse probability treatment weighting (IPTW) and a doubly robust estimator. We evaluate our approach by refuting the outcome. Finally, we find that social media usage has a significant impact on spending patterns.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130875832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-18DOI: 10.1109/ICCIT54785.2021.9689821
Mahtab Alam, M. Raihan, Mubtasim Rafid Chowdhury, A. Shams
Electrooculography (EOG) signal is the potential difference between the cornea and the retina of the eye. The voltage amplitude changes when the eye moves in various directions. This change produces a distinct EOG pattern when the eye moves in a particular direction. Therefore, by monitoring the EOG signal, it is possible to track the eye movement. The EOG based eye-tracking technique can be extended to maneuver smart wheelchairs for neurodegenerative disease patients. For a successful operation of such a smart wheelchair, an accurate classification of the EOG signal is required. In this experimental study, we collected two channel EOG signals in the laboratory from multiple individuals and propose an Artificial Neural Network (ANN) based method to differentiate among the nine classes of EOG signals: up, down, left, right, down-left, down-right, up-left, up-right, and blink. This wide range classification would be suitable to perform complicated tasks in smart technology platform. Our model can successfully predict the eye movement from the statistical properties and dominant frequency of the measured EOG signal with an accuracy, precision, recall, and F1 score of 99%. This is a significant improvement over past studies conducted by various researchers for the same purpose and to the knowledge of the authors, such a high accuracy has not been previously achieved for the nine classes of EOG signals mentioned earlier. The proposed model is compatible for real-time smart applications based on eye movements.
{"title":"High Precision Eye Tracking Based on Electrooculography (EOG) Signal Using Artificial Neural Network (ANN) for Smart Technology Application","authors":"Mahtab Alam, M. Raihan, Mubtasim Rafid Chowdhury, A. Shams","doi":"10.1109/ICCIT54785.2021.9689821","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689821","url":null,"abstract":"Electrooculography (EOG) signal is the potential difference between the cornea and the retina of the eye. The voltage amplitude changes when the eye moves in various directions. This change produces a distinct EOG pattern when the eye moves in a particular direction. Therefore, by monitoring the EOG signal, it is possible to track the eye movement. The EOG based eye-tracking technique can be extended to maneuver smart wheelchairs for neurodegenerative disease patients. For a successful operation of such a smart wheelchair, an accurate classification of the EOG signal is required. In this experimental study, we collected two channel EOG signals in the laboratory from multiple individuals and propose an Artificial Neural Network (ANN) based method to differentiate among the nine classes of EOG signals: up, down, left, right, down-left, down-right, up-left, up-right, and blink. This wide range classification would be suitable to perform complicated tasks in smart technology platform. Our model can successfully predict the eye movement from the statistical properties and dominant frequency of the measured EOG signal with an accuracy, precision, recall, and F1 score of 99%. This is a significant improvement over past studies conducted by various researchers for the same purpose and to the knowledge of the authors, such a high accuracy has not been previously achieved for the nine classes of EOG signals mentioned earlier. The proposed model is compatible for real-time smart applications based on eye movements.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122215661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Diabetic Retinopathy (DR) is a kind of diabetes complication that damages the light-sensitive tissues of the blood vessels at the back of the eyes. Early detection of such problems along with controlling diabetes can prevent severe damages from the disease. Detection of DR is time-consuming, and manual detection is error-prone. Hence, in the majority of the cases, it is detected at a severe stage making it difficult to treat properly. To handle this problem, this paper presents a deep learning model consisting of AlexNet, VGGNet, and modified VGGNet, and ResNet, to detect DR from images. A detailed comparison among the adopted models and the state-of-the-art reveals that the modified VGGNet outperforms other applied models with 87.69% accuracy, 87.93% precision, and 87.81% recall. The model accuracy increases to 95.77% after performing hyperparameter tuning. The experimental results are promising and make the model a suitable candidate for automated DR detection from fundus images.
{"title":"An Improved Diabetic Retinopathy Image Classification by Using Deep Learning Models","authors":"Jannatul Naim, Zahid Hasan, Md. Niajul Haque Pradhan, Shamim Ripon","doi":"10.1109/ICCIT54785.2021.9689874","DOIUrl":"https://doi.org/10.1109/ICCIT54785.2021.9689874","url":null,"abstract":"Diabetic Retinopathy (DR) is a kind of diabetes complication that damages the light-sensitive tissues of the blood vessels at the back of the eyes. Early detection of such problems along with controlling diabetes can prevent severe damages from the disease. Detection of DR is time-consuming, and manual detection is error-prone. Hence, in the majority of the cases, it is detected at a severe stage making it difficult to treat properly. To handle this problem, this paper presents a deep learning model consisting of AlexNet, VGGNet, and modified VGGNet, and ResNet, to detect DR from images. A detailed comparison among the adopted models and the state-of-the-art reveals that the modified VGGNet outperforms other applied models with 87.69% accuracy, 87.93% precision, and 87.81% recall. The model accuracy increases to 95.77% after performing hyperparameter tuning. The experimental results are promising and make the model a suitable candidate for automated DR detection from fundus images.","PeriodicalId":166450,"journal":{"name":"2021 24th International Conference on Computer and Information Technology (ICCIT)","volume":"5 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122603131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：