Pub Date : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893691
M. Akhand, M. K. Das, N. Siddique
The COVID-19 pandemic has been a challenging time that the mankind had experienced since the Spanish flue where there is no available treatment except supportive care. The patients with COVID-19 suffered from mild to severe breathing difficulties and respiratory support was the main reason of hospitalization. Ventilator is generally used for the respiratory support which mixes air under pressure with required oxygen concentrations. Invasive mechanical ventilator (IMV) is a complex computer-driven machine delivering positive pressure to the lungs via an endotracheal or tracheostomy tube to support full ventilation. IMV is very expensive and the operation requires specialist nurses. An alternative to IMV is a non-invasive ventilation (NIV) which was deemed necessary during the pandemic. Continuous positive airway pressure (CPAP) is a NIV applied through a face mask and does not require specialist nurses. Due to low cost and simple operation, CPAP drew attention during COVID-19 pandemic. This paper presents the design and development of a CPAP ventilation device. The designed CPAP is a microcontroller based electro-mechanical device for supportive care of patients with respiratory problem.
{"title":"Design and Development of a Robust CPAP Device for Respiratory Support","authors":"M. Akhand, M. K. Das, N. Siddique","doi":"10.1109/BECITHCON54710.2021.9893691","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893691","url":null,"abstract":"The COVID-19 pandemic has been a challenging time that the mankind had experienced since the Spanish flue where there is no available treatment except supportive care. The patients with COVID-19 suffered from mild to severe breathing difficulties and respiratory support was the main reason of hospitalization. Ventilator is generally used for the respiratory support which mixes air under pressure with required oxygen concentrations. Invasive mechanical ventilator (IMV) is a complex computer-driven machine delivering positive pressure to the lungs via an endotracheal or tracheostomy tube to support full ventilation. IMV is very expensive and the operation requires specialist nurses. An alternative to IMV is a non-invasive ventilation (NIV) which was deemed necessary during the pandemic. Continuous positive airway pressure (CPAP) is a NIV applied through a face mask and does not require specialist nurses. Due to low cost and simple operation, CPAP drew attention during COVID-19 pandemic. This paper presents the design and development of a CPAP ventilation device. The designed CPAP is a microcontroller based electro-mechanical device for supportive care of patients with respiratory problem.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"412 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123370409","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893681
F. Ahmed, Shams Nafisa Ali, Tanzila Akter, J. Ferdous
This study aims to analyze the impact of the structure of a cryoprobe tip on the transient thermal phenomenon that takes place during cryoablation of breast cancer. A 2D axisymmetric breast model with an embedded tumor has been constructed using COMSOL Multiphysics® interface. Three different tip structures (conical, spherical, cylindrical) have been considered for the experimentation to determine the optimal shape which not only destroys a greater fraction of tumor volume with great rapidity but also ensures minimal damage to the neighboring healthy tissues. Fine triangular meshes have been generated all over the experimentation domain. The simulation has been performed for a duration of 200s employing Pennes bioheat equation with relevant thermo-physical properties of tissue layers and appropriate boundary conditions as well as initial conditions for each of the structure. From the result illustrated via the frozen fraction vs time plot, it can be deduced that despite having nearly same and comparable dimensions, the cylindrical probe tip, outperforms the conical and spherical tips by a margin of 13.57% and 7.44%, respectively in terms of destroying the tumor tissue volume. Therefore, the result shows that the probe tip with a greater surface area demonstrates better cryogenic activity which is in conformity with the expectation. In conclusion, the study delivers a significant insight for manufacturing especially engineered cryoprobe tip with adequate proof of concept and ushers a new pathway for enhancing the net efficacy of the cryosurgical intervention for the treatment of breast cancer. Still, the study offers scopes for further optimization to make it more realistic, effective and clinically relevant.
{"title":"Evaluating the Impact of Cryoprobe Tip Structure for Effective Cryoablation of Breast Cancer","authors":"F. Ahmed, Shams Nafisa Ali, Tanzila Akter, J. Ferdous","doi":"10.1109/BECITHCON54710.2021.9893681","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893681","url":null,"abstract":"This study aims to analyze the impact of the structure of a cryoprobe tip on the transient thermal phenomenon that takes place during cryoablation of breast cancer. A 2D axisymmetric breast model with an embedded tumor has been constructed using COMSOL Multiphysics® interface. Three different tip structures (conical, spherical, cylindrical) have been considered for the experimentation to determine the optimal shape which not only destroys a greater fraction of tumor volume with great rapidity but also ensures minimal damage to the neighboring healthy tissues. Fine triangular meshes have been generated all over the experimentation domain. The simulation has been performed for a duration of 200s employing Pennes bioheat equation with relevant thermo-physical properties of tissue layers and appropriate boundary conditions as well as initial conditions for each of the structure. From the result illustrated via the frozen fraction vs time plot, it can be deduced that despite having nearly same and comparable dimensions, the cylindrical probe tip, outperforms the conical and spherical tips by a margin of 13.57% and 7.44%, respectively in terms of destroying the tumor tissue volume. Therefore, the result shows that the probe tip with a greater surface area demonstrates better cryogenic activity which is in conformity with the expectation. In conclusion, the study delivers a significant insight for manufacturing especially engineered cryoprobe tip with adequate proof of concept and ushers a new pathway for enhancing the net efficacy of the cryosurgical intervention for the treatment of breast cancer. Still, the study offers scopes for further optimization to make it more realistic, effective and clinically relevant.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"192 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123538365","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893641
I. Hussain, Md. Azam Hossain, Se-Jin Park
Neurological impairment is a common disorder observed in stroke population and Electroencephalography (EEG) monitoring is considered a significant marker for diagnostics stroke onset. This study aims to propose a proof-of-concept of a healthcare “digital twin” and utilize EEG data and machine-learning models to build a digital twin for the stroke patients. We examined 48 stroke patients admitted to a rehabilitation clinic and 75 healthy persons. Portable EEG devices were used to capture EEG using frontal, central, temporal, and occipital cortical electrodes. The statistical analysis revealed that the revised brain-symmetry index, theta, and delta activities are relevant characteristics for classifying stroke patients and healthy individuals in motor and cognitive states. Using the machine learning approach, Support vector machine (SVM) classified the EEG feature dataset with 76% accuracy (AUC: 0.84) for classifying the stroke and the control group. This healthcare digital twin framework may assist in clinical decision-making for stroke preventive measures and post-stroke treatment.
{"title":"A Healthcare Digital Twin for Diagnosis of Stroke","authors":"I. Hussain, Md. Azam Hossain, Se-Jin Park","doi":"10.1109/BECITHCON54710.2021.9893641","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893641","url":null,"abstract":"Neurological impairment is a common disorder observed in stroke population and Electroencephalography (EEG) monitoring is considered a significant marker for diagnostics stroke onset. This study aims to propose a proof-of-concept of a healthcare “digital twin” and utilize EEG data and machine-learning models to build a digital twin for the stroke patients. We examined 48 stroke patients admitted to a rehabilitation clinic and 75 healthy persons. Portable EEG devices were used to capture EEG using frontal, central, temporal, and occipital cortical electrodes. The statistical analysis revealed that the revised brain-symmetry index, theta, and delta activities are relevant characteristics for classifying stroke patients and healthy individuals in motor and cognitive states. Using the machine learning approach, Support vector machine (SVM) classified the EEG feature dataset with 76% accuracy (AUC: 0.84) for classifying the stroke and the control group. This healthcare digital twin framework may assist in clinical decision-making for stroke preventive measures and post-stroke treatment.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114641980","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893623
Ggaliwango Marvin, Md. Golam Rabiul Alarm
The global trends of women in the reproductive age have significantly altered due to their personal and career development engagements besides adoption of contraceptive methods. Since women are extending birth to their late ages where natural conception is quite hard besides other factors, it has globally boosted the fertility service market which is a projected 41.4 billion industry by 2026. Despite the growing market for fertility services, infertility evaluation is still uncomfortable, expensive, inaccessible and ambiguous for both the customers and the fertility service providers. In this work, we deploy Machine Learning and Explainable Artificial Intelligence to predict the outcomes of fertility treatment using interpretable Machine Learning Lattice Models for predictive, preventive and precision reproductive medicine. We also introduce the concept of Quantum Lattice Learning in Artificial Intelligence for Machine Learning Interpretability.
{"title":"An Explainable Lattice based Fertility Treatment Outcome Prediction Model for TeleFertility","authors":"Ggaliwango Marvin, Md. Golam Rabiul Alarm","doi":"10.1109/BECITHCON54710.2021.9893623","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893623","url":null,"abstract":"The global trends of women in the reproductive age have significantly altered due to their personal and career development engagements besides adoption of contraceptive methods. Since women are extending birth to their late ages where natural conception is quite hard besides other factors, it has globally boosted the fertility service market which is a projected 41.4 billion industry by 2026. Despite the growing market for fertility services, infertility evaluation is still uncomfortable, expensive, inaccessible and ambiguous for both the customers and the fertility service providers. In this work, we deploy Machine Learning and Explainable Artificial Intelligence to predict the outcomes of fertility treatment using interpretable Machine Learning Lattice Models for predictive, preventive and precision reproductive medicine. We also introduce the concept of Quantum Lattice Learning in Artificial Intelligence for Machine Learning Interpretability.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129550932","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893583
Sangita Roy, S. S. Chaudhuri
Adverse climate conditions affect digital photography causing colour shifting, poor visibility, contrast reduction, and fainted appearance due to the scattering of atmospheric Particulate Matter (APM). To get an optimum transmission matrix is the key success of any single image dehazing technique. Deep Learning based Super Resolution technique with VDSR 20-weighted Layers ImageNet classifier improves any image resolution leading to noise suppression. High Residual Learning gradient clipping makes the algorithm converge fast with denoising and removal of artifacts by compression. This key observation has been exercised in improving resolution of the hazy images with an optical image formation model. In addition, benchmark established images are evaluated and their comparisons to the state-of-the-art methods show a consistent improvement in accurate scene transmission estimation resulting in clear, natural haze-free radiance. A good balance between execution speed and processing speed has been achieved.
{"title":"Single Image Very Deep Super Resolution (SIVDSR) Dehaze","authors":"Sangita Roy, S. S. Chaudhuri","doi":"10.1109/BECITHCON54710.2021.9893583","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893583","url":null,"abstract":"Adverse climate conditions affect digital photography causing colour shifting, poor visibility, contrast reduction, and fainted appearance due to the scattering of atmospheric Particulate Matter (APM). To get an optimum transmission matrix is the key success of any single image dehazing technique. Deep Learning based Super Resolution technique with VDSR 20-weighted Layers ImageNet classifier improves any image resolution leading to noise suppression. High Residual Learning gradient clipping makes the algorithm converge fast with denoising and removal of artifacts by compression. This key observation has been exercised in improving resolution of the hazy images with an optical image formation model. In addition, benchmark established images are evaluated and their comparisons to the state-of-the-art methods show a consistent improvement in accurate scene transmission estimation resulting in clear, natural haze-free radiance. A good balance between execution speed and processing speed has been achieved.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122727885","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893719
Ggaliwango Marvin, Md. Golam Rabiul Alam
Neonatal Intensive Care Units (NICU) service costs are rapidly growing due to the higher resource utilization intensity. This in turn increases the healthcare costs for NICU patients besides the inaccessibility and unpreparedness of both NICU service providers and patient caretakers hence an increase in neonatal mortality and morbidity. There a lot of contributors to NICU admissions but the exiting methods consider very limited features to precisely predict NICU admissions. In this paper, we present a robust Explainable Artificial Intelligence approach that allows machines to interpretably learn from a pool of possible contributing features in order to predict an NICU admission. Our machine learning approach interpretably illustrates the thought process of admission prediction to the physician and patient. This provides transparent and trustable insights for the precise, proactive, personalized and participatory NICU medical diagnostics and treatment plans for the patient. We statistically and visually present Random Forest and Logistic Regression prediction explanations using SHAP, LIME and ELI5 techniques. This predictive technological approach can preventively increase success of maternal and neonatal monitoring and treatment plans. It can also enhance proactive management of NICU facilities (resources) by the responsible facility administrators most especially in resource constrained settings.
{"title":"Explainable Feature Learning for Predicting Neonatal Intensive Care Unit (NICU) Admissions","authors":"Ggaliwango Marvin, Md. Golam Rabiul Alam","doi":"10.1109/BECITHCON54710.2021.9893719","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893719","url":null,"abstract":"Neonatal Intensive Care Units (NICU) service costs are rapidly growing due to the higher resource utilization intensity. This in turn increases the healthcare costs for NICU patients besides the inaccessibility and unpreparedness of both NICU service providers and patient caretakers hence an increase in neonatal mortality and morbidity. There a lot of contributors to NICU admissions but the exiting methods consider very limited features to precisely predict NICU admissions. In this paper, we present a robust Explainable Artificial Intelligence approach that allows machines to interpretably learn from a pool of possible contributing features in order to predict an NICU admission. Our machine learning approach interpretably illustrates the thought process of admission prediction to the physician and patient. This provides transparent and trustable insights for the precise, proactive, personalized and participatory NICU medical diagnostics and treatment plans for the patient. We statistically and visually present Random Forest and Logistic Regression prediction explanations using SHAP, LIME and ELI5 techniques. This predictive technological approach can preventively increase success of maternal and neonatal monitoring and treatment plans. It can also enhance proactive management of NICU facilities (resources) by the responsible facility administrators most especially in resource constrained settings.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125685738","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893548
Zahrul Jannat Peya, M. Ferdous, M. Akhand, Mohammed Golam Zilani, N. Siddique
Autism or Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder marked by repetitive and characteristic patterns of behavior as well as social communication and interaction impairments. Nowadays ASD is a great concern worldwide and its detection is an important issue for the better treatment. As ASD is neurodevelopmental disorder, brain signals play, especially electroencephalography (EEG), is shown potential sources for ASD detection. There are different approaches for ASD detection with processing and/or transforming EEG signals. This study investigated ASD detection employing fractal dimension measure on EEG data. Higuchi’s Fractal Dimension (HFD) is measured in resting-state eyes-closed EEG recording of 25 subjects. It is identified that HFD is sensitive to the brain activity and ASD detection is possible from HFD values.
{"title":"ASD Detection using Higuchi’s Fractal Dimension from EEG","authors":"Zahrul Jannat Peya, M. Ferdous, M. Akhand, Mohammed Golam Zilani, N. Siddique","doi":"10.1109/BECITHCON54710.2021.9893548","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893548","url":null,"abstract":"Autism or Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder marked by repetitive and characteristic patterns of behavior as well as social communication and interaction impairments. Nowadays ASD is a great concern worldwide and its detection is an important issue for the better treatment. As ASD is neurodevelopmental disorder, brain signals play, especially electroencephalography (EEG), is shown potential sources for ASD detection. There are different approaches for ASD detection with processing and/or transforming EEG signals. This study investigated ASD detection employing fractal dimension measure on EEG data. Higuchi’s Fractal Dimension (HFD) is measured in resting-state eyes-closed EEG recording of 25 subjects. It is identified that HFD is sensitive to the brain activity and ASD detection is possible from HFD values.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123743536","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 : 2021-12-04DOI: 10.1109/BECITHCON54710.2021.9893578
Annesha Ahsan, Nazmun Nessa Moon, Shayla Sharmin, Mohammad Monirul Islam, Refath Ara Hossain, Samia Nawshin
At present day, fatal road accidents have become a very common fact all over the world and also in Bangladesh. It is increasing day by day in big cities like Dhaka. Thousands of lives are taken every year due to traffic accidents. In this research paper, we have tried to justify the cause behind fatal traffic accidents. By taking several causes as attributes such as the age of driver behind the wheel, experience, vehicle types, health issues of the driver, and so on. Using these causes as the main input criteria we took data records from various fatal accident cases and also non-fatal accident cases through news sources and surveys. In consideration of our research, we applied machine learning algorithms like Decision trees, Random Forest Classifier which justifies our proposed model accuracy. Through the data mining technique, we have got a satisfactory percentage of accuracy of about 95% for Decision Tree Classifier and 93% for Random Forest Classifier.
{"title":"Machine Learning Approach to Predict Traffic Accident Occurrence in Bangladesh","authors":"Annesha Ahsan, Nazmun Nessa Moon, Shayla Sharmin, Mohammad Monirul Islam, Refath Ara Hossain, Samia Nawshin","doi":"10.1109/BECITHCON54710.2021.9893578","DOIUrl":"https://doi.org/10.1109/BECITHCON54710.2021.9893578","url":null,"abstract":"At present day, fatal road accidents have become a very common fact all over the world and also in Bangladesh. It is increasing day by day in big cities like Dhaka. Thousands of lives are taken every year due to traffic accidents. In this research paper, we have tried to justify the cause behind fatal traffic accidents. By taking several causes as attributes such as the age of driver behind the wheel, experience, vehicle types, health issues of the driver, and so on. Using these causes as the main input criteria we took data records from various fatal accident cases and also non-fatal accident cases through news sources and surveys. In consideration of our research, we applied machine learning algorithms like Decision trees, Random Forest Classifier which justifies our proposed model accuracy. Through the data mining technique, we have got a satisfactory percentage of accuracy of about 95% for Decision Tree Classifier and 93% for Random Forest Classifier.","PeriodicalId":170083,"journal":{"name":"2021 IEEE International Conference on Biomedical Engineering, Computer and Information Technology for Health (BECITHCON)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133127466","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：