Pub Date : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319429
Alireza Rezaie Zangene, Ali Abbasi
The purpose of this research was to continuous knee joint angle estimation from sEMG during squat using artificial neural networks. sEMG signals of vastus medialis, rectus femoris, biceps femoris and 3D kinematics of lower extremity joints for four participants during squat were captured at 1500 Hz and 100 Hz, respectively. sEMG signals were preprocessed and RMS and variance were extracted as input features. The processed input data was given to a three-layer feed forward neural network with one hidden layer. The proposed network was trained by the Levenberg-Marquardt algorithm. The root mean square error (RMSE) and correlation coefficient (CC) were used to evaluate the accuracy of estimation. The results showed that this network is able to continuously estimate the knee joint angle with global RMSE of 5.0041° ± 0.9963° and CC of 0.9898 ± 0.0039. It concludes that a multilayer neural network with a simple structure has the ability to continuously estimate the joint angle from sEMG data while performing an athletic movement under real loading situation.
{"title":"Continuous Estimation of Knee Joint Angle during Squat from sEMG using Artificial Neural Networks","authors":"Alireza Rezaie Zangene, Ali Abbasi","doi":"10.1109/ICBME51989.2020.9319429","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319429","url":null,"abstract":"The purpose of this research was to continuous knee joint angle estimation from sEMG during squat using artificial neural networks. sEMG signals of vastus medialis, rectus femoris, biceps femoris and 3D kinematics of lower extremity joints for four participants during squat were captured at 1500 Hz and 100 Hz, respectively. sEMG signals were preprocessed and RMS and variance were extracted as input features. The processed input data was given to a three-layer feed forward neural network with one hidden layer. The proposed network was trained by the Levenberg-Marquardt algorithm. The root mean square error (RMSE) and correlation coefficient (CC) were used to evaluate the accuracy of estimation. The results showed that this network is able to continuously estimate the knee joint angle with global RMSE of 5.0041° ± 0.9963° and CC of 0.9898 ± 0.0039. It concludes that a multilayer neural network with a simple structure has the ability to continuously estimate the joint angle from sEMG data while performing an athletic movement under real loading situation.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122225379","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319422
Mahdi Heydari, A. Nazari, A. Tanbakoosaz
The aim of this study was to compare the kinematic coordination of the lower extremity between elite and professional athletes during Roundhouse kick using a modified vector coding technique. For this purpose, 20 wushu sanda athletes (10 elite and 10 professionals) participated in the study voluntarily. The kinematic data were recorded by lower limb plug in gait biomechanical model and the 12-camera VICON motion capture system. After processing the data, the coordination of the lower limb joints on the sagittal plane was calculated using the vector coding method in MATLAB software. Independent t-test was used for statistical analysis. The findings showed that in ankle and knee coordination, antiphase pattern with more ankle dominancy, in knee and hip coordination, antiphase pattern with knee dominancy, and in hip and pelvic coordination, antiphase pattern with hip dominancy increased in elite athletes. The results of this study showed that the elite athletes, in performing the Roundhouse kick as an open kinematics chain movement, performed the movement transition from the upper joint to the lower joint more optimally. Also, they can support lower joint with a movement reduction in the upper one and perform the movement effectively.
{"title":"Comparison of kinematics coordination of lower extremity between Elit and Professional athletes during roundhouse kick using modified vector coding technique","authors":"Mahdi Heydari, A. Nazari, A. Tanbakoosaz","doi":"10.1109/ICBME51989.2020.9319422","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319422","url":null,"abstract":"The aim of this study was to compare the kinematic coordination of the lower extremity between elite and professional athletes during Roundhouse kick using a modified vector coding technique. For this purpose, 20 wushu sanda athletes (10 elite and 10 professionals) participated in the study voluntarily. The kinematic data were recorded by lower limb plug in gait biomechanical model and the 12-camera VICON motion capture system. After processing the data, the coordination of the lower limb joints on the sagittal plane was calculated using the vector coding method in MATLAB software. Independent t-test was used for statistical analysis. The findings showed that in ankle and knee coordination, antiphase pattern with more ankle dominancy, in knee and hip coordination, antiphase pattern with knee dominancy, and in hip and pelvic coordination, antiphase pattern with hip dominancy increased in elite athletes. The results of this study showed that the elite athletes, in performing the Roundhouse kick as an open kinematics chain movement, performed the movement transition from the upper joint to the lower joint more optimally. Also, they can support lower joint with a movement reduction in the upper one and perform the movement effectively.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127159755","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319446
Nasim Rahmanifar, F. Eskandari, M. Shafieian
Helmet is the most effective personal protective equipment for motorcyclists that can significantly reduce the risk of skull fracture and traumatic brain injuries. The performance of the helmets is evaluated using a drop-weight impact tower test. Due to restrictions to access the drop tower test assembly in many low-income countries, evaluation of helmet performance is a major challenge. The current study aimed to develop a simple biomechanical method to help the assessment of motorcycle helmets without employing sophisticated test setups. Since liner foam is the main component of the helmets that can absorb the impact energy, the viscoelastic properties of expanded polystyrene (EPS), as a most commonly used foam in the helmets, were characterized. A simple lumped model of the head and the helmet was developed to characterize the head linear acceleration under impact condition. The results of the current study showed that the value of head injury criteria (HIC) was less than the maximum acceptable value for the standard. Also, the results of scanning electron microscopy (SEM) showed that EPS foam which has experienced 10% compressive strain could recover after loading; indicating that the motorcycle helmet contained EPS as a liner foam could pass the safety standards.
{"title":"Mechanical Characterization of Expanded Polystyrene (EPS) as a Liner Foam in Motorcycle Helmets","authors":"Nasim Rahmanifar, F. Eskandari, M. Shafieian","doi":"10.1109/ICBME51989.2020.9319446","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319446","url":null,"abstract":"Helmet is the most effective personal protective equipment for motorcyclists that can significantly reduce the risk of skull fracture and traumatic brain injuries. The performance of the helmets is evaluated using a drop-weight impact tower test. Due to restrictions to access the drop tower test assembly in many low-income countries, evaluation of helmet performance is a major challenge. The current study aimed to develop a simple biomechanical method to help the assessment of motorcycle helmets without employing sophisticated test setups. Since liner foam is the main component of the helmets that can absorb the impact energy, the viscoelastic properties of expanded polystyrene (EPS), as a most commonly used foam in the helmets, were characterized. A simple lumped model of the head and the helmet was developed to characterize the head linear acceleration under impact condition. The results of the current study showed that the value of head injury criteria (HIC) was less than the maximum acceptable value for the standard. Also, the results of scanning electron microscopy (SEM) showed that EPS foam which has experienced 10% compressive strain could recover after loading; indicating that the motorcycle helmet contained EPS as a liner foam could pass the safety standards.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127343733","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319432
Mohsen Annabestani, M. Sayad, Pouria Esmaeili-Dokht, Razieh Gorji, M. Fardmanesh
one of the most critical disadvantages of Ionic Polymer Metal Composites (IPMCs) is their Back Relaxation (BR) effect. The BR is an unwanted and slow counter-bending of IPMCs that shows itself in the opposite direction of the desired bending. Some techniques based on using closed-loop control have been proposed for eliminating the BR effect. However, those techniques are valid only for small deformation of IPMC, and also they use closed-loop approaches. While in practical applications, the IPMC deformation should be large, and we cannot use the closed-loop system. To address these problems, we have proposed a non-feedback method for restraining the IPMC BR effect, which works even in large bending displacement. The proposed technique is based on fast reciprocating motions of free water molecules in the Nafion membrane which is producing by exogenous noise stimulations on the pattern-free electrodes of IPMC. Our idea has been described conceptually and validated by several experiments.
{"title":"Eliminating Back Relaxation in Large-Deformable IPMC Artificial Muscles: A Noise-Assistive Pattern-Free Electrode Approach","authors":"Mohsen Annabestani, M. Sayad, Pouria Esmaeili-Dokht, Razieh Gorji, M. Fardmanesh","doi":"10.1109/ICBME51989.2020.9319432","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319432","url":null,"abstract":"one of the most critical disadvantages of Ionic Polymer Metal Composites (IPMCs) is their Back Relaxation (BR) effect. The BR is an unwanted and slow counter-bending of IPMCs that shows itself in the opposite direction of the desired bending. Some techniques based on using closed-loop control have been proposed for eliminating the BR effect. However, those techniques are valid only for small deformation of IPMC, and also they use closed-loop approaches. While in practical applications, the IPMC deformation should be large, and we cannot use the closed-loop system. To address these problems, we have proposed a non-feedback method for restraining the IPMC BR effect, which works even in large bending displacement. The proposed technique is based on fast reciprocating motions of free water molecules in the Nafion membrane which is producing by exogenous noise stimulations on the pattern-free electrodes of IPMC. Our idea has been described conceptually and validated by several experiments.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121382964","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319420
Marjan Khaledi, M. Dehghani, Mohsen Mohammadi, Roozbeh Abolpour
In recent years, heart transplantation has been considered a suitable method for heart failure (HF), a significant cardiovascular disease. Heart failure often means chronic heart failure or congestive heart failure. In this case, the heart is unable to pump enough blood to the vital organs. Usually, heart failure occurs on the left side of the heart that pumps blood throughout the body. No one can ever deny the impact of a left ventricular assist device (LVAD) as a bridge for recovery in the patient’s treatment process with congestive heart failure. In this paper, we aim to control the LVAD flow and the goal is to return the person to everyday life by using this assist device and control it perfectly. Since heart failure could occur if the pump speed is lower or higher than some thresholds, to prevent heart failure, the controller should provide an appropriate pump speed. This paper prohibits these ranges by designing a suitable controller that can track the flow of healthy heart behavior through a Frank-Starling-Like controller. The controller performance is evaluated by a simulated model of the heart and the results are satisfactory.
{"title":"Controller Design for Left Ventricular Assist Devices in Patients with Heart Failure","authors":"Marjan Khaledi, M. Dehghani, Mohsen Mohammadi, Roozbeh Abolpour","doi":"10.1109/ICBME51989.2020.9319420","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319420","url":null,"abstract":"In recent years, heart transplantation has been considered a suitable method for heart failure (HF), a significant cardiovascular disease. Heart failure often means chronic heart failure or congestive heart failure. In this case, the heart is unable to pump enough blood to the vital organs. Usually, heart failure occurs on the left side of the heart that pumps blood throughout the body. No one can ever deny the impact of a left ventricular assist device (LVAD) as a bridge for recovery in the patient’s treatment process with congestive heart failure. In this paper, we aim to control the LVAD flow and the goal is to return the person to everyday life by using this assist device and control it perfectly. Since heart failure could occur if the pump speed is lower or higher than some thresholds, to prevent heart failure, the controller should provide an appropriate pump speed. This paper prohibits these ranges by designing a suitable controller that can track the flow of healthy heart behavior through a Frank-Starling-Like controller. The controller performance is evaluated by a simulated model of the heart and the results are satisfactory.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116603597","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319453
Negar Massihi, S. Rashidi
Nowadays, biometric systems play a main role in personal information protection. Each person has a distinct pattern of wrist veins that can be used for authentication. Fractional Fourier transform (FrFT) changes signal to complex form. In this paper, FrFT was utilized for the time-frequency analysis of images. After the pre-processing stage and extracting veins from the background, the phase of FrFT coefficients was computed for each image. The PUT database was used in the proposed method for verifying individuals. This database consists of 1200 wrist and 1200 palm vein images. In this paper, wrist vein images were only utilized. Receiver Operating Characteristic (ROC) and Support Vector Machine (SVM) were used for feature selection and classification, respectively. The results showed that wrist vein images can be used for verification and FrFT is a capable tool for feature extraction as the average accuracy was obtained 99.65±0.95% in the operating point.
{"title":"Authentication of Individuals based on Wrist Vein Images by Extracting Phase of Fractional Fourier Transform","authors":"Negar Massihi, S. Rashidi","doi":"10.1109/ICBME51989.2020.9319453","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319453","url":null,"abstract":"Nowadays, biometric systems play a main role in personal information protection. Each person has a distinct pattern of wrist veins that can be used for authentication. Fractional Fourier transform (FrFT) changes signal to complex form. In this paper, FrFT was utilized for the time-frequency analysis of images. After the pre-processing stage and extracting veins from the background, the phase of FrFT coefficients was computed for each image. The PUT database was used in the proposed method for verifying individuals. This database consists of 1200 wrist and 1200 palm vein images. In this paper, wrist vein images were only utilized. Receiver Operating Characteristic (ROC) and Support Vector Machine (SVM) were used for feature selection and classification, respectively. The results showed that wrist vein images can be used for verification and FrFT is a capable tool for feature extraction as the average accuracy was obtained 99.65±0.95% in the operating point.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"278 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126015727","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319450
Homayoon Soleimani Dinani, Mehrab Pourmadadi, H. Rashedi, F. Yazdian
An electrochemical nanobiosensor is described for the voltammetric determination of miRNA-128 as the cancer biomarker. Aptamer chains were immobilized on the surface of a glassy carbon electrode (GCE) via gold nanoparticles/magnetite/reduced graphene oxide (AuNPs/Fe3O4/RGO). FTIR and XRD analysis were used for the characterization of synthesized nanomaterials. Square wave voltammetry (SWV) was used to characterize the modified GCE in a label-free method. The results indicate that the modified working electrode has high selectivity and for miRNA-128 over other biomolecules. The hexacyanoferrate redox system, typically operated at around 0.3 V (vs. Ag/AgCl) was used as an electrochemical probe. The limit of detection (LOD) and linear detection range are 0.05346 fM and 0.1 to 0.9 fM, respectively.
{"title":"Fabrication of nanomaterial-based biosensor for measurement of a microRNA involved in cancer","authors":"Homayoon Soleimani Dinani, Mehrab Pourmadadi, H. Rashedi, F. Yazdian","doi":"10.1109/ICBME51989.2020.9319450","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319450","url":null,"abstract":"An electrochemical nanobiosensor is described for the voltammetric determination of miRNA-128 as the cancer biomarker. Aptamer chains were immobilized on the surface of a glassy carbon electrode (GCE) via gold nanoparticles/magnetite/reduced graphene oxide (AuNPs/Fe3O4/RGO). FTIR and XRD analysis were used for the characterization of synthesized nanomaterials. Square wave voltammetry (SWV) was used to characterize the modified GCE in a label-free method. The results indicate that the modified working electrode has high selectivity and for miRNA-128 over other biomolecules. The hexacyanoferrate redox system, typically operated at around 0.3 V (vs. Ag/AgCl) was used as an electrochemical probe. The limit of detection (LOD) and linear detection range are 0.05346 fM and 0.1 to 0.9 fM, respectively.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125533105","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319457
R. Movahed, M. Rezaeian, Sina Javadifar, Mohammadreza Alimoradijazi
Today, face recognition systems play a crucial role in many access control and automatic identification systems. However, these systems still have shortcomings that reduce their performance efficiency. In this paper, a novel face recognition framework is introduced, combining the Eigenfaces algorithm and image registration. Firstly, the collected face images are preprocessed, then the Eigenfaces algorithm is applied to them for obtaining the reference eigenvectors. After that, three test images are captured using a webcam, and the images' faces are detected using the Viola-jones algorithm. The detected faces are registered to the collected face images, and the detected face with the lowest mean square error is selected for subsequent steps. Next, the selected detected face's eigenvector and the distance between it and reference eigenvectors are calculated, respectively. The minimum distance is then compared with a manual threshold to recognize the person as an unknown or known person. If the person is recognized as a known person, the person's identity is identified as the person belongs to the minimum distance. For validating the presented method, a public and an exclusive face image database are used. The obtained results indicate that the proposed framework achieved a better performance than traditional similarity-based methods to recognize known and unknown persons and identify known persons.
{"title":"A Face Recognition Framework Based on the Integration of Eigenfaces Algorithm and Image Registration Technique","authors":"R. Movahed, M. Rezaeian, Sina Javadifar, Mohammadreza Alimoradijazi","doi":"10.1109/ICBME51989.2020.9319457","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319457","url":null,"abstract":"Today, face recognition systems play a crucial role in many access control and automatic identification systems. However, these systems still have shortcomings that reduce their performance efficiency. In this paper, a novel face recognition framework is introduced, combining the Eigenfaces algorithm and image registration. Firstly, the collected face images are preprocessed, then the Eigenfaces algorithm is applied to them for obtaining the reference eigenvectors. After that, three test images are captured using a webcam, and the images' faces are detected using the Viola-jones algorithm. The detected faces are registered to the collected face images, and the detected face with the lowest mean square error is selected for subsequent steps. Next, the selected detected face's eigenvector and the distance between it and reference eigenvectors are calculated, respectively. The minimum distance is then compared with a manual threshold to recognize the person as an unknown or known person. If the person is recognized as a known person, the person's identity is identified as the person belongs to the minimum distance. For validating the presented method, a public and an exclusive face image database are used. The obtained results indicate that the proposed framework achieved a better performance than traditional similarity-based methods to recognize known and unknown persons and identify known persons.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121594463","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319426
K. Rezaee, Afsoon Badiei, S. Meshgini
As a contagious disease originating from a novel coronavirus, COVID-19 leads to swollen air sacs in the lungs. It can be diagnosed using a chest X-ray (CXR) images, which is usually cheaper and less harmful than a CT scan and is always available in small or rural hospitals. X-ray machines, however, sometimes cannot diagnose COVID-19. Since the COVID-19 dataset is small and cannot be diagnosed from CXR, pre-trained neural networks can be employed for coronavirus diagnosis. This paper mainly aims to use pre-trained deep transfer learning (DTL) architectures and conventional machine learning (ML) models as an automated instrument to diagnose COVID-19 from CXRs. To overcome the lack of a large number of images, DTL is utilized to extract image features for better classification. Then, to optimize the decision-making level for infectious diseases similar to bacterial and viral pneumonia, the extracted features are selected and classified. Our proposed method was validated by creating a new CXR database from Vasei Hospital in Sabzevar, Iran. Our hybrid model achieved hit rates above 99% and outperformed for CXR of COVID-19 and similar pneumonia classification. Comparative analysis shows the superiority of the proposed COVID-19 classification model based on DTL over other competitive methods.
{"title":"A hybrid deep transfer learning based approach for COVID-19 classification in chest X-ray images","authors":"K. Rezaee, Afsoon Badiei, S. Meshgini","doi":"10.1109/ICBME51989.2020.9319426","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319426","url":null,"abstract":"As a contagious disease originating from a novel coronavirus, COVID-19 leads to swollen air sacs in the lungs. It can be diagnosed using a chest X-ray (CXR) images, which is usually cheaper and less harmful than a CT scan and is always available in small or rural hospitals. X-ray machines, however, sometimes cannot diagnose COVID-19. Since the COVID-19 dataset is small and cannot be diagnosed from CXR, pre-trained neural networks can be employed for coronavirus diagnosis. This paper mainly aims to use pre-trained deep transfer learning (DTL) architectures and conventional machine learning (ML) models as an automated instrument to diagnose COVID-19 from CXRs. To overcome the lack of a large number of images, DTL is utilized to extract image features for better classification. Then, to optimize the decision-making level for infectious diseases similar to bacterial and viral pneumonia, the extracted features are selected and classified. Our proposed method was validated by creating a new CXR database from Vasei Hospital in Sabzevar, Iran. Our hybrid model achieved hit rates above 99% and outperformed for CXR of COVID-19 and similar pneumonia classification. Comparative analysis shows the superiority of the proposed COVID-19 classification model based on DTL over other competitive methods.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126421189","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 : 2020-11-26DOI: 10.1109/ICBME51989.2020.9319449
F. Shanehsazzadeh, Shahin Rouhi, Tala Ahmadvand, Mahrokh Namazi, S. Kiani, M. Fardmanesh
In this paper, a novel, low-cost multi-electrode array fabrication method has been proposed for epidural spinal cord stimulation (ESCS) to restore motion ability after paralysis due to spinal cord injury (SCI). In this approach, unlike the existing neural prosthesis technologies, costly standard microfabrication processes are eliminated. This makes the proposed electrodes cost-effective and suitable for massive production for clinical applications. Based on the conformability and mechanical compliance of the electrodes and spinal cord tissue, full polydimethylsiloxane (PDMS)-based passive multi-electrode array structure is proposed. The suggested neural electrode array structure consists of PDMS (as substrate) and embedded 100 μm-thick Cu wires (as conductive parts). Despite the fabrication and implementation challenges caused by the low Young modulus of PDMS, its suitable mechanical properties close to those of the spinal cord tissues make PDMS one of the best options. These electrodes were used along with pulse generator circuits providing biphasic pulse waveforms for two channels. Employing the ESCS system caused movement in the paralyzed limbs in an adult male Wistar rat. Our preliminary ESCS studies have shown that such electrode arrays are capable of neuronal stimulation effectively.
{"title":"A Novel, Low Cost and Versatile Fabrication Method of Flexible Multi-electrode Array for Spinal Cord Stimulation","authors":"F. Shanehsazzadeh, Shahin Rouhi, Tala Ahmadvand, Mahrokh Namazi, S. Kiani, M. Fardmanesh","doi":"10.1109/ICBME51989.2020.9319449","DOIUrl":"https://doi.org/10.1109/ICBME51989.2020.9319449","url":null,"abstract":"In this paper, a novel, low-cost multi-electrode array fabrication method has been proposed for epidural spinal cord stimulation (ESCS) to restore motion ability after paralysis due to spinal cord injury (SCI). In this approach, unlike the existing neural prosthesis technologies, costly standard microfabrication processes are eliminated. This makes the proposed electrodes cost-effective and suitable for massive production for clinical applications. Based on the conformability and mechanical compliance of the electrodes and spinal cord tissue, full polydimethylsiloxane (PDMS)-based passive multi-electrode array structure is proposed. The suggested neural electrode array structure consists of PDMS (as substrate) and embedded 100 μm-thick Cu wires (as conductive parts). Despite the fabrication and implementation challenges caused by the low Young modulus of PDMS, its suitable mechanical properties close to those of the spinal cord tissues make PDMS one of the best options. These electrodes were used along with pulse generator circuits providing biphasic pulse waveforms for two channels. Employing the ESCS system caused movement in the paralyzed limbs in an adult male Wistar rat. Our preliminary ESCS studies have shown that such electrode arrays are capable of neuronal stimulation effectively.","PeriodicalId":120969,"journal":{"name":"2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127655149","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：