Pub Date : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101598
Nirupam Das, Md. Selim Hossain
In this research, the dependency of radar target returns from an ocean surface on different sea conditions is investigated, and found that sea reflectivity has a strong dependency on sea conditions. We have seen that, sea surface reflectivity is proportionally correlated with frequency and marine roughness. The dependency of sea reflectivity on polarization and grazing angle is further investigated and found that when radar frequency is increased, the dependency on polarization is decreased, and while increasing grazing angle the dependency on it is decreased. We further investigated that for the same roughness and the variety of grazing angles taken, horizontally polarized transmissions have a lower reflectance than vertically polarized transmissions.
{"title":"Investigation of the Impact of Sea Conditions on the Sea Surface Reflectivity in Maritime Radar Sea Clutter Modeling","authors":"Nirupam Das, Md. Selim Hossain","doi":"10.1109/ECCE57851.2023.10101598","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101598","url":null,"abstract":"In this research, the dependency of radar target returns from an ocean surface on different sea conditions is investigated, and found that sea reflectivity has a strong dependency on sea conditions. We have seen that, sea surface reflectivity is proportionally correlated with frequency and marine roughness. The dependency of sea reflectivity on polarization and grazing angle is further investigated and found that when radar frequency is increased, the dependency on polarization is decreased, and while increasing grazing angle the dependency on it is decreased. We further investigated that for the same roughness and the variety of grazing angles taken, horizontally polarized transmissions have a lower reflectance than vertically polarized transmissions.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131303442","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101653
Laboni Paul, K. H. Talukder
Type-II diabetes is growing at an alarming rate worldwide. Eventually, it leads to visual impairment by damaging the retinal blood vessels, which is known as Diabetic Retinopathy (DR). A plethora of research is ongoing to automate the early detection of DR to help the doctor with periodic eye examinations due to advancements in AI based computer vision algorithms and camera technology. In this paper, we have proposed a comparison between two widely used very deep convolutional neural network architectures (ResNet-101 v2, InceptionResNet V2) with the comparatively new optimized and composite scalable architecture (EfficientNet B5) with while training them from scratch and using them as a pre-trained to transfer learning. We have found out the pre-trained EfficientNet B5 has outperformed our other candidates as well as available methods in the current literature by achieving accuracy of 97.78%. We also provide detailed enough information to make the result reproducible.
{"title":"Blindness Risk Prediction caused by Diabetic Retinopathy from Retinal Image","authors":"Laboni Paul, K. H. Talukder","doi":"10.1109/ECCE57851.2023.10101653","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101653","url":null,"abstract":"Type-II diabetes is growing at an alarming rate worldwide. Eventually, it leads to visual impairment by damaging the retinal blood vessels, which is known as Diabetic Retinopathy (DR). A plethora of research is ongoing to automate the early detection of DR to help the doctor with periodic eye examinations due to advancements in AI based computer vision algorithms and camera technology. In this paper, we have proposed a comparison between two widely used very deep convolutional neural network architectures (ResNet-101 v2, InceptionResNet V2) with the comparatively new optimized and composite scalable architecture (EfficientNet B5) with while training them from scratch and using them as a pre-trained to transfer learning. We have found out the pre-trained EfficientNet B5 has outperformed our other candidates as well as available methods in the current literature by achieving accuracy of 97.78%. We also provide detailed enough information to make the result reproducible.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124608609","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101548
Md. Rabiul Hasan, Shah Muhammad Azmat Ullah, Md. Ebtidaul Karim
Pneumonia is now a life-threatening respiratory illness that can affect the lungs. Mainly the aged and children suffer the most. If the right diagnosis is not made, it could be fatal. So early diagnosis is very much needed to save many human lives. For diagnosis purposes Medical imaging, such as a chest x-ray can be utilized effectively and skilled radiologists are needed for this. Due to the blurriness of X-ray images, proper diagnosis can be difficult and time-consuming, even for radiographers with experience. As human judgment is involved, a pneumonia diagnosis may be erroneous. Hence, a deep learning-based automated system can be used to assist the radiographer in taking decisions more precisely and accurately. There have been several existing methods available for diagnosing pneumonia but they have accuracy issues. In this paper, we seek to automate the process of identifying and categorizing cases of pneumonia from CXR images deploying deep CNN. A deep CNN model has been built from scratch which will automate the process and provide high diagnosis performance. After passing through multiple convolutional layers and corresponding max pooling layers, the information is then fed into the dense layers. Lastly, using the sigmoidal function, the classification is performed. The model's performance improves as it simultaneously gains training and reduces loss.
{"title":"An Effective Framework for Identifying Pneumonia in Healthcare Using a Convolutional Neural Network","authors":"Md. Rabiul Hasan, Shah Muhammad Azmat Ullah, Md. Ebtidaul Karim","doi":"10.1109/ECCE57851.2023.10101548","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101548","url":null,"abstract":"Pneumonia is now a life-threatening respiratory illness that can affect the lungs. Mainly the aged and children suffer the most. If the right diagnosis is not made, it could be fatal. So early diagnosis is very much needed to save many human lives. For diagnosis purposes Medical imaging, such as a chest x-ray can be utilized effectively and skilled radiologists are needed for this. Due to the blurriness of X-ray images, proper diagnosis can be difficult and time-consuming, even for radiographers with experience. As human judgment is involved, a pneumonia diagnosis may be erroneous. Hence, a deep learning-based automated system can be used to assist the radiographer in taking decisions more precisely and accurately. There have been several existing methods available for diagnosing pneumonia but they have accuracy issues. In this paper, we seek to automate the process of identifying and categorizing cases of pneumonia from CXR images deploying deep CNN. A deep CNN model has been built from scratch which will automate the process and provide high diagnosis performance. After passing through multiple convolutional layers and corresponding max pooling layers, the information is then fed into the dense layers. Lastly, using the sigmoidal function, the classification is performed. The model's performance improves as it simultaneously gains training and reduces loss.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131213372","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101521
Salvir Hossain, Md Tawabur Rahman
The measurement of pH is an important routine practice in many chemical and biomedical applications. This work reports an Ion Sensitive Field Effect Transistors (ISFET) based pH sensor. The two-dimensional modeling of the sensor is performed in the COMSOL Multiphysics® v. 6.0 platform using its semiconductor module, electrostatics module, and transport of diluted species module. The binding of ions in Si02 results in induced charge carriers in the conducting channel of ISFET, which is controlled by the applied gate voltage for determining ion concentration. Here, the pH of water as the bulk electrolyte is measured by attaining the required gate voltage to achieve a certain drain current in ISFET. The sensor shows excellent sensitivities of 48.7 mV/pH and 41.3 mV/pH with linear detection ranges of pH 1–7 and 8–13, respectively. The excellent sensitivity and wide linear detection range can be attributed to the high concentration of surface sites in the Si02 sensing film and improved disassociation constants in the presence of the gate oxide in contact with the electrolyte. Finally, this sensor demonstrates its potential for real applications.
{"title":"COMSOL-Based Modeling and Simulation of ISFET pH Sensor Using Si02 Sensing Film","authors":"Salvir Hossain, Md Tawabur Rahman","doi":"10.1109/ECCE57851.2023.10101521","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101521","url":null,"abstract":"The measurement of pH is an important routine practice in many chemical and biomedical applications. This work reports an Ion Sensitive Field Effect Transistors (ISFET) based pH sensor. The two-dimensional modeling of the sensor is performed in the COMSOL Multiphysics® v. 6.0 platform using its semiconductor module, electrostatics module, and transport of diluted species module. The binding of ions in Si02 results in induced charge carriers in the conducting channel of ISFET, which is controlled by the applied gate voltage for determining ion concentration. Here, the pH of water as the bulk electrolyte is measured by attaining the required gate voltage to achieve a certain drain current in ISFET. The sensor shows excellent sensitivities of 48.7 mV/pH and 41.3 mV/pH with linear detection ranges of pH 1–7 and 8–13, respectively. The excellent sensitivity and wide linear detection range can be attributed to the high concentration of surface sites in the Si02 sensing film and improved disassociation constants in the presence of the gate oxide in contact with the electrolyte. Finally, this sensor demonstrates its potential for real applications.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132133761","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10100744
M. Hossain, M. A. Kashem, Shabnom Mustary
Agricultural yield generally depends on the level of soil fertility. Nitrogen (N), Phosphorus (P), Potassium (K), pH, the temperature of the soil, and moisture as soil chemical constituents are fundamental parameters for determining soil fertility. Good yield can easily be ensured by measuring their presence and applying the right amount of fertilizer in the right season. Most farmers do not produce good crops due to insufficient knowledge and the inability to use the proper amount of fertilizers. Current methods of measuring soil nutrients involve collecting soil from the field and transporting it to a laboratory for testing, which is often subjective and very expensive. This paper suggests an efficient IoT-based soil nutrient monitoring and machine learning-based crop recommendation system that helps farmers by offering crop-related details and recommendations for crops based on different soil and weather attributes. The proposed system deploys various types of sensors to determine soil nutrients, these sensors continuously collect the required data from the farm field and transmit it via a wireless sensor network (WSN) to a cloud database. By monitoring (N, P, K, temperature, pH, humidity, rainfall) values and analyzing the permanent and temporary behavior of the soil, the machine learning approach will recommend what types of crops have the best production potential for this land. Agriculture's use of machine-learning technology makes it easier to select the best-yielding crops by reducing the cost of unnecessary fertilizer use, which reduces manual labor in crop and crop management and increases productivity. The most appropriate crops for that cropland are suggested using machine learning algorithms in IoT-based soil nutrient monitoring, which stores data from various soil nutrients in a database. As a result, agricultural production will contribute more to national growth.
{"title":"IoT Based Smart Soil Fertilizer Monitoring And ML Based Crop Recommendation System","authors":"M. Hossain, M. A. Kashem, Shabnom Mustary","doi":"10.1109/ECCE57851.2023.10100744","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10100744","url":null,"abstract":"Agricultural yield generally depends on the level of soil fertility. Nitrogen (N), Phosphorus (P), Potassium (K), pH, the temperature of the soil, and moisture as soil chemical constituents are fundamental parameters for determining soil fertility. Good yield can easily be ensured by measuring their presence and applying the right amount of fertilizer in the right season. Most farmers do not produce good crops due to insufficient knowledge and the inability to use the proper amount of fertilizers. Current methods of measuring soil nutrients involve collecting soil from the field and transporting it to a laboratory for testing, which is often subjective and very expensive. This paper suggests an efficient IoT-based soil nutrient monitoring and machine learning-based crop recommendation system that helps farmers by offering crop-related details and recommendations for crops based on different soil and weather attributes. The proposed system deploys various types of sensors to determine soil nutrients, these sensors continuously collect the required data from the farm field and transmit it via a wireless sensor network (WSN) to a cloud database. By monitoring (N, P, K, temperature, pH, humidity, rainfall) values and analyzing the permanent and temporary behavior of the soil, the machine learning approach will recommend what types of crops have the best production potential for this land. Agriculture's use of machine-learning technology makes it easier to select the best-yielding crops by reducing the cost of unnecessary fertilizer use, which reduces manual labor in crop and crop management and increases productivity. The most appropriate crops for that cropland are suggested using machine learning algorithms in IoT-based soil nutrient monitoring, which stores data from various soil nutrients in a database. As a result, agricultural production will contribute more to national growth.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129532560","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101589
Sabbir A. Rahman, N. Sharmin, Md. Mahbubur Rahman
A tropical cyclone is one of the most egregious natural disasters in the world that brings calamity to coastal lives by hitting the corresponding country's bordering basins since ancient time. The rapid intensification of TC has always been a threat to the coastal peoples living in different corners of the world. Geographical locations and geographical settings of being a low-lying deltaic country could trigger this calamitous event and bring individual hazards like a storm surge, inundation, oceanic flood, and many more. Tracking a tropical cyclone is not an easy task as it shows nonlinear behavior to different models to forecast. However, considering several limitations, experts from different countries use several products like satellite images, numerical data, and radar images to predict the formation, track, and the intensity of a cyclone. However, it is concerning that a full-fledged automatic cyclone prediction visualization tool for the wider populace does not exist. In this work, we are unlikely to provide an absolute automated visualization tool. Rather, we attempted to compensate for the lack of one by creating a prototype of a cyclone prediction and visualization dashboard with Streamlit, a Python framework for rapidly developing machine learning web apps. Furthermore, we considered visualizing the data sets in order to interpret them from various perspectives, and we used optical flow to determine the cyclonic behaviors as another approach.
{"title":"Cyclone Prediction Visualization Tools Using Machine Learning Models and Optical Flow","authors":"Sabbir A. Rahman, N. Sharmin, Md. Mahbubur Rahman","doi":"10.1109/ECCE57851.2023.10101589","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101589","url":null,"abstract":"A tropical cyclone is one of the most egregious natural disasters in the world that brings calamity to coastal lives by hitting the corresponding country's bordering basins since ancient time. The rapid intensification of TC has always been a threat to the coastal peoples living in different corners of the world. Geographical locations and geographical settings of being a low-lying deltaic country could trigger this calamitous event and bring individual hazards like a storm surge, inundation, oceanic flood, and many more. Tracking a tropical cyclone is not an easy task as it shows nonlinear behavior to different models to forecast. However, considering several limitations, experts from different countries use several products like satellite images, numerical data, and radar images to predict the formation, track, and the intensity of a cyclone. However, it is concerning that a full-fledged automatic cyclone prediction visualization tool for the wider populace does not exist. In this work, we are unlikely to provide an absolute automated visualization tool. Rather, we attempted to compensate for the lack of one by creating a prototype of a cyclone prediction and visualization dashboard with Streamlit, a Python framework for rapidly developing machine learning web apps. Furthermore, we considered visualizing the data sets in order to interpret them from various perspectives, and we used optical flow to determine the cyclonic behaviors as another approach.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":" 33","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113952017","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101522
Mujahidul Islam, Maqsudur Rahman, M. T. Ahmed, Abu Zafor Muhammad Islam, Dipankar Das, M. M. Hoque
Harassment is a kind of act that annoys or upsets someone. Harassment can be classified into different categories. Sexual harassment is one of them. Sexual harassment is a type of harassment that involves the use of implicit or explicit sexual overtones, including the inappropriate and unwelcome promises of rewards in exchange for sexual favors. At present time, the technology has become more advance and spread all over the place. That gave the toxic people a huge opportunity to spread toxicity in online platforms. Because of the increasing amount Bangla text in different social media platforms, we also need to filter such kinds of offensive Bangla texts. The objective of this research is to detect sexual harassment from Bangla text and classify them by using machine learning and deep learning algorithms as well as prevents them. In the experiment, we combined TF-IDF with different machine learning algorithms like Naive Bayes, Decision Tree, Random Forest, AdaBoost, SGD, Logistic Regression, KNN, SVM and got accuracy of 74.9%, 75.6%, 70.0%, 70.1%, 75.2%, 75.7%, 65.2%, 76.5% respectively. Deep learning algorithms like CNN, LSTM, hybrid CNN-LSTM were also used and achieved accuracy of 89% for all of them which is comparatively better than machine learning techniques.
{"title":"Sexual Harassment Detection using Machine Learning and Deep Learning Techniques for Bangla Text","authors":"Mujahidul Islam, Maqsudur Rahman, M. T. Ahmed, Abu Zafor Muhammad Islam, Dipankar Das, M. M. Hoque","doi":"10.1109/ECCE57851.2023.10101522","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101522","url":null,"abstract":"Harassment is a kind of act that annoys or upsets someone. Harassment can be classified into different categories. Sexual harassment is one of them. Sexual harassment is a type of harassment that involves the use of implicit or explicit sexual overtones, including the inappropriate and unwelcome promises of rewards in exchange for sexual favors. At present time, the technology has become more advance and spread all over the place. That gave the toxic people a huge opportunity to spread toxicity in online platforms. Because of the increasing amount Bangla text in different social media platforms, we also need to filter such kinds of offensive Bangla texts. The objective of this research is to detect sexual harassment from Bangla text and classify them by using machine learning and deep learning algorithms as well as prevents them. In the experiment, we combined TF-IDF with different machine learning algorithms like Naive Bayes, Decision Tree, Random Forest, AdaBoost, SGD, Logistic Regression, KNN, SVM and got accuracy of 74.9%, 75.6%, 70.0%, 70.1%, 75.2%, 75.7%, 65.2%, 76.5% respectively. Deep learning algorithms like CNN, LSTM, hybrid CNN-LSTM were also used and achieved accuracy of 89% for all of them which is comparatively better than machine learning techniques.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116039701","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101579
Md. Safaiat Hossain, Rafiul Bari Angan, M. Hasan
This study focuses on the real-world application of a state-of-the-art convolution neural network (CNN) based pothole detection model and its practical implications. A multiple-scenario analysis was performed, combining several experiments based on video recorded at different environmental conditions and vehicle speeds. Moreover, the performance of the CNN-based YOLOv4-tiny AI model was compared with an expert human grader (civil engineer). The findings from the comparative analysis suggest that out of five different cases, the AI-based model (69.57-85.00%) outperformed the human evaluator (43.67-80.67%) in four cases, with the highest accuracy of 85%. This indicates the utility of using an AI-based approach to pothole detection, especially in regional areas of developing countries like Bangladesh.
{"title":"Pothole Detection and Estimation of Repair Cost in Bangladeshi Street: AI-based Multiple Case Analysis","authors":"Md. Safaiat Hossain, Rafiul Bari Angan, M. Hasan","doi":"10.1109/ECCE57851.2023.10101579","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101579","url":null,"abstract":"This study focuses on the real-world application of a state-of-the-art convolution neural network (CNN) based pothole detection model and its practical implications. A multiple-scenario analysis was performed, combining several experiments based on video recorded at different environmental conditions and vehicle speeds. Moreover, the performance of the CNN-based YOLOv4-tiny AI model was compared with an expert human grader (civil engineer). The findings from the comparative analysis suggest that out of five different cases, the AI-based model (69.57-85.00%) outperformed the human evaluator (43.67-80.67%) in four cases, with the highest accuracy of 85%. This indicates the utility of using an AI-based approach to pothole detection, especially in regional areas of developing countries like Bangladesh.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116690367","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}
Cardiovascular diseases (CVDs), which include heart disorders, are the most prevalent and significant causes of death worldwide, including Bangladesh. Blood artery problems, rhythm issues, chest pain, heart attacks, strokes, and erratic blood pressure are a few of these. In Bangladesh, cardiovascular disease is the main factor in both male and female fatalities. More than 80% of CVD deaths are caused by heart disease and strokes, which are the predominant causes. To be able to examine the effectiveness of the various models, this research article explains the underlying methods as Support vector machines (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), and XGBoost (XGB), wherein Random Forest perform better when their hyperparameters are tuned (RandomizedSearchCV). There suggested ensemble technique such as Bagging, Voting, Stacking. Additionally, it is suggested that a hybrid strategy using Bagging and stacking ensemble approaches can boost the predictability of cardiovascular disease. For this analysis of patient performance, we used a dataset from Kaggle that comprises of 70,000 unique data values. According to the experiment's findings, the proposed model had the best disease prediction accuracy, coming in at 84.03%.
{"title":"An Improved Framework for Reliable Cardiovascular Disease Prediction Using Hybrid Ensemble Learning","authors":"Tanjim Mahmud, Anik Barua, M. Begum, Eipshita Chakma, Sudhakar Das, Nahed Sharmen","doi":"10.1109/ECCE57851.2023.10101564","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101564","url":null,"abstract":"Cardiovascular diseases (CVDs), which include heart disorders, are the most prevalent and significant causes of death worldwide, including Bangladesh. Blood artery problems, rhythm issues, chest pain, heart attacks, strokes, and erratic blood pressure are a few of these. In Bangladesh, cardiovascular disease is the main factor in both male and female fatalities. More than 80% of CVD deaths are caused by heart disease and strokes, which are the predominant causes. To be able to examine the effectiveness of the various models, this research article explains the underlying methods as Support vector machines (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), and XGBoost (XGB), wherein Random Forest perform better when their hyperparameters are tuned (RandomizedSearchCV). There suggested ensemble technique such as Bagging, Voting, Stacking. Additionally, it is suggested that a hybrid strategy using Bagging and stacking ensemble approaches can boost the predictability of cardiovascular disease. For this analysis of patient performance, we used a dataset from Kaggle that comprises of 70,000 unique data values. According to the experiment's findings, the proposed model had the best disease prediction accuracy, coming in at 84.03%.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114650007","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101536
Aditta Chowdhury, Diba Das, R. Cheung, M. Chowdhury
This paper aims to design a digital system to pre-process electrocardiogram (ECG) and photoplethysmogram (PPG) signal for the purpose of hardware implementation. Muscle signal, motion artifacts, power line interference affect the biomedical signal during data acquisition. The proposed system focuses at removing the noises by designing infinite impulse response filter to remove power line noise and finite impulse response filter to eliminate other high and low frequency noises. At first the preprocessor is designed in Matlab to validate the simulation performance. Then the hardware is designed in xilinx system generator targeting Zedboard Zynq xc7z020-1clg484. Finally, we verified the hardware software codesign by comparing both outputs. For quantity based analysis different filtering techniques have been applied to determine the most optimized system in terms of resource utilization and power consumption. Pearson correlation coefficient of 0.9993 and 0.9982 have been found for ECG and PPG, respectively using Hamming filter technique for High and low pass filter. Root squared error for both signal has been also in the range of 10−2• These data validate the accuracy of the designed system providing quality assurance. Frequency spectrum also has been analyzed to ensure denoising of undesired signals. The designed preprocessor can be utilized for further analysis of the signals and designing digital systems & wearable devices for the detection of heart rate, cardiac diseases etc.
{"title":"Hardware/Software Co-design of an ECG- PPG Preprocessor: A Qualitative & Quantitative Analysis","authors":"Aditta Chowdhury, Diba Das, R. Cheung, M. Chowdhury","doi":"10.1109/ECCE57851.2023.10101536","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101536","url":null,"abstract":"This paper aims to design a digital system to pre-process electrocardiogram (ECG) and photoplethysmogram (PPG) signal for the purpose of hardware implementation. Muscle signal, motion artifacts, power line interference affect the biomedical signal during data acquisition. The proposed system focuses at removing the noises by designing infinite impulse response filter to remove power line noise and finite impulse response filter to eliminate other high and low frequency noises. At first the preprocessor is designed in Matlab to validate the simulation performance. Then the hardware is designed in xilinx system generator targeting Zedboard Zynq xc7z020-1clg484. Finally, we verified the hardware software codesign by comparing both outputs. For quantity based analysis different filtering techniques have been applied to determine the most optimized system in terms of resource utilization and power consumption. Pearson correlation coefficient of 0.9993 and 0.9982 have been found for ECG and PPG, respectively using Hamming filter technique for High and low pass filter. Root squared error for both signal has been also in the range of 10−2• These data validate the accuracy of the designed system providing quality assurance. Frequency spectrum also has been analyzed to ensure denoising of undesired signals. The designed preprocessor can be utilized for further analysis of the signals and designing digital systems & wearable devices for the detection of heart rate, cardiac diseases etc.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131908907","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：