Pub Date : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677751
Ni Wayan Yulya Wiani, A. Arifin, M. Fatoni, Josaphat Pramudijanto
Stroke is a potentially fatal illness caused by clotting of the blood vessels that supply oxygen to the brain. Up to 65 percent of stroke patients are affected by Hemiparesis. Muscle weakness is a typical side effect, which might lead to a reduction in physical activity. This makes it difficult for post-stroke patients to carry out daily tasks. Therefore, a game-based rehabilitation strategy focused on grasping movement is recommended to help the upper limbs recover. Individual biomedical signals were used to control the game. EMG instrumentation used to process biomedical signals. To aid in this process, hand gloves are also used to evaluate the range of motion produced during rehabilitation. The game becomes more exciting by using Leap Motion to track patient hand movements and move virtual hands in the game. The experimental results revealed an average increase in the amplitude of the LEMG signal generated by participants 1 and 2. The average amplitude increase in subject 1 was 22.81 mV, while it was 89.60 mV in subject 2. For further research, a compact and sensitive EMG instrumentation can be built. In addition, real-time computing can be used to build rehabilitation systems that can detect the onset of LEMG and create more interactive games.
{"title":"Instrumentation Design of Game Rehabilitation with Myoelectric Command","authors":"Ni Wayan Yulya Wiani, A. Arifin, M. Fatoni, Josaphat Pramudijanto","doi":"10.1109/BioSMART54244.2021.9677751","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677751","url":null,"abstract":"Stroke is a potentially fatal illness caused by clotting of the blood vessels that supply oxygen to the brain. Up to 65 percent of stroke patients are affected by Hemiparesis. Muscle weakness is a typical side effect, which might lead to a reduction in physical activity. This makes it difficult for post-stroke patients to carry out daily tasks. Therefore, a game-based rehabilitation strategy focused on grasping movement is recommended to help the upper limbs recover. Individual biomedical signals were used to control the game. EMG instrumentation used to process biomedical signals. To aid in this process, hand gloves are also used to evaluate the range of motion produced during rehabilitation. The game becomes more exciting by using Leap Motion to track patient hand movements and move virtual hands in the game. The experimental results revealed an average increase in the amplitude of the LEMG signal generated by participants 1 and 2. The average amplitude increase in subject 1 was 22.81 mV, while it was 89.60 mV in subject 2. For further research, a compact and sensitive EMG instrumentation can be built. In addition, real-time computing can be used to build rehabilitation systems that can detect the onset of LEMG and create more interactive games.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121321497","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677841
Dibyo Fabian Dofadar, Riyo Hayat Khan, Md. Golam Rabiul Alam
The number of people affected by Coronavirus is quite concerning in Bangladesh. It has become a necessity to forecast the future cases since it involves ensuring adequate resources to help people and imposing strict guidelines to deal with this epidemic. This research is about predicting upcoming COVID-19 confirmed cases and deaths from a time series dataset using Hidden Markov Model. The optimal number of hidden states were determined using AIC and BIC. The proposed models are implemented to forecast the daily confirmed cases and daily deaths of Bangladesh for next 90 days.
{"title":"COVID-19 Confirmed Cases and Deaths Prediction in Bangladesh Using Hidden Markov Model","authors":"Dibyo Fabian Dofadar, Riyo Hayat Khan, Md. Golam Rabiul Alam","doi":"10.1109/BioSMART54244.2021.9677841","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677841","url":null,"abstract":"The number of people affected by Coronavirus is quite concerning in Bangladesh. It has become a necessity to forecast the future cases since it involves ensuring adequate resources to help people and imposing strict guidelines to deal with this epidemic. This research is about predicting upcoming COVID-19 confirmed cases and deaths from a time series dataset using Hidden Markov Model. The optimal number of hidden states were determined using AIC and BIC. The proposed models are implemented to forecast the daily confirmed cases and daily deaths of Bangladesh for next 90 days.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115845098","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677831
Bingbin Wang, E. Kamavuako
The long-term robustness of pattern recognition-based myoelectric systems draws more attention from researchers. Though, there is a lack of analysis investigating how features change over time. This study used two metrics: Coefficient of variation of the first four moments (CoV) and Two-Sample Kolmogorov-Smirnov Test statistics (K-S); to quantify the stability of feature distributions and correlate their changes over time to classification performance. We acquired two surface electromyography (sEMG) channels from sixteen subjects (ten able-bodied and six trans-radial amputees) performing three hand motions. Results showed that the selected metrics correlate to some degree to classification accuracy. Feature distributions are affected less by the time when data are combined. These results imply that stable temporal change may be an acceptable way to choose robust features in long term investigations.
{"title":"Correlation between the stability of feature distribution and classification performance in sEMG signals","authors":"Bingbin Wang, E. Kamavuako","doi":"10.1109/BioSMART54244.2021.9677831","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677831","url":null,"abstract":"The long-term robustness of pattern recognition-based myoelectric systems draws more attention from researchers. Though, there is a lack of analysis investigating how features change over time. This study used two metrics: Coefficient of variation of the first four moments (CoV) and Two-Sample Kolmogorov-Smirnov Test statistics (K-S); to quantify the stability of feature distributions and correlate their changes over time to classification performance. We acquired two surface electromyography (sEMG) channels from sixteen subjects (ten able-bodied and six trans-radial amputees) performing three hand motions. Results showed that the selected metrics correlate to some degree to classification accuracy. Feature distributions are affected less by the time when data are combined. These results imply that stable temporal change may be an acceptable way to choose robust features in long term investigations.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115174971","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677680
R. P. Fraga, Ziho Kang, Junehyung Lee, J. Crutchfield
A dynamic task refers to a task in which the state of the system can dynamically change when a user interacts with the system's components. For example, when an air traffic controller detects aircraft on converging flight paths, the controller can select from multiple altitude, heading and speed clearances to maintain safe separation between them. Some clearance options for those two aircraft, however, may lead to losses of separation with other aircraft. One viable non-intrusive approach to characterize a user's interaction with a system is through real-time analysis of eye movements at the time when the state of the system is changing. The presentation of data from such analyses could be an effective way to enhance user training techniques. In this article, we provide a framework of how to analyze eye-tracking data to identify useful characteristics along with associated algorithms, followed by a simple case study to validate our framework.
{"title":"Real-time eye tracking analysis for training in a dynamic task","authors":"R. P. Fraga, Ziho Kang, Junehyung Lee, J. Crutchfield","doi":"10.1109/BioSMART54244.2021.9677680","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677680","url":null,"abstract":"A dynamic task refers to a task in which the state of the system can dynamically change when a user interacts with the system's components. For example, when an air traffic controller detects aircraft on converging flight paths, the controller can select from multiple altitude, heading and speed clearances to maintain safe separation between them. Some clearance options for those two aircraft, however, may lead to losses of separation with other aircraft. One viable non-intrusive approach to characterize a user's interaction with a system is through real-time analysis of eye movements at the time when the state of the system is changing. The presentation of data from such analyses could be an effective way to enhance user training techniques. In this article, we provide a framework of how to analyze eye-tracking data to identify useful characteristics along with associated algorithms, followed by a simple case study to validate our framework.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121771124","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677808
Guochang Ye, Han Deng, C. Woodworth, Mehmet Kaya
Cell differentiation is a progressive process and hard to quantitate without advanced biotechnological methods. In this study, a machine learning (ML) algorithm is introduced to detect the undifferentiated cell clusters and improve time and labor efficiencies by clustering image features extracted from the changing morphology of immortalized cervical cells. The methodology involves taking phase-contrast image data from the monolayer cell culture of the human cervical epithelial cell. The normalized histogram features and Haralick texture features from each dividing tile of input images are used in a simple k-means clustering training. The resulting colored maps are generated by filling each tile with a specific color according to its classification label. The targeted color representing the undifferentiation is selected automatically. Then simple image processing techniques are applied to analyze the colored map and outline the contour of undifferentiated cell clusters on the input images. The results showed that the undifferentiated cell clusters are indicated clearly in the images. After visually comparing to the ground truth cell morphology, the proposed method could accurately pinpoint the major undifferentiated cell clusters with minimal costs.
{"title":"An Unsupervised Machine Learning Algorithm to Detect Undifferentiated Cell Clusters of Immortalized Human Cervical Epithelial Cell","authors":"Guochang Ye, Han Deng, C. Woodworth, Mehmet Kaya","doi":"10.1109/BioSMART54244.2021.9677808","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677808","url":null,"abstract":"Cell differentiation is a progressive process and hard to quantitate without advanced biotechnological methods. In this study, a machine learning (ML) algorithm is introduced to detect the undifferentiated cell clusters and improve time and labor efficiencies by clustering image features extracted from the changing morphology of immortalized cervical cells. The methodology involves taking phase-contrast image data from the monolayer cell culture of the human cervical epithelial cell. The normalized histogram features and Haralick texture features from each dividing tile of input images are used in a simple k-means clustering training. The resulting colored maps are generated by filling each tile with a specific color according to its classification label. The targeted color representing the undifferentiation is selected automatically. Then simple image processing techniques are applied to analyze the colored map and outline the contour of undifferentiated cell clusters on the input images. The results showed that the undifferentiated cell clusters are indicated clearly in the images. After visually comparing to the ground truth cell morphology, the proposed method could accurately pinpoint the major undifferentiated cell clusters with minimal costs.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129987941","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677778
Mohammed Mohammed, K. Elleithy, Wafa Elmannai
KMSAFE APP is a location-based safety smartphone application. This application allows users like students, faculty, and staff to send an emergency notification to the campus security department by using a single small button connected by Bluetooth to their smartphone. They can keep this button in their key chain, attached to their pants, or keep it in their backpack. They just need to push the button for 3 seconds; the campus security department will receive a real-time emergency notification from the person in danger. This message will include the victim's personal information, map position, and images of the surrounding area, including the criminal. Our results showed that KMSAFE APP was able to help security personnel respond to emergencies is 81 % faster than traditional mobile applications.
{"title":"KMSAFE APP: Campus Safety Mobile App","authors":"Mohammed Mohammed, K. Elleithy, Wafa Elmannai","doi":"10.1109/BioSMART54244.2021.9677778","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677778","url":null,"abstract":"KMSAFE APP is a location-based safety smartphone application. This application allows users like students, faculty, and staff to send an emergency notification to the campus security department by using a single small button connected by Bluetooth to their smartphone. They can keep this button in their key chain, attached to their pants, or keep it in their backpack. They just need to push the button for 3 seconds; the campus security department will receive a real-time emergency notification from the person in danger. This message will include the victim's personal information, map position, and images of the surrounding area, including the criminal. Our results showed that KMSAFE APP was able to help security personnel respond to emergencies is 81 % faster than traditional mobile applications.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127089926","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677889
F. Samann, T. Schanze
Noise reduction of considerable recorded data, e.g., EEG, PPG signals, is significantly important in biomedical signal processing. Singular value decomposition (SVD) method has shown optimistic results in denoising biomedical dataset of images and signals via dimension reduction. However, a still challenge in SVD approach is to find the low-rank representation of the matrix obtained by matricification of the signal of interest adaptively which retrain the energy in signal subspace and neglect the energy in noise subspace. Here, we develop an adaptive rank estimation by the SVD for denoising purpose based on estimating the noise level σest using the first level detail symmlet-wavelet's coefficients d1. The optimal rank is obtained at the point where the difference between the noisy and the reduced rank dataset is approximately below the estimated noise level. The proposed method has successfully estimated the optimal rank which gives the best denoising performance.
{"title":"Denoising biomedical signals via adaptive low-rank matrix representation by singular value decomposition using wavelets","authors":"F. Samann, T. Schanze","doi":"10.1109/BioSMART54244.2021.9677889","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677889","url":null,"abstract":"Noise reduction of considerable recorded data, e.g., EEG, PPG signals, is significantly important in biomedical signal processing. Singular value decomposition (SVD) method has shown optimistic results in denoising biomedical dataset of images and signals via dimension reduction. However, a still challenge in SVD approach is to find the low-rank representation of the matrix obtained by matricification of the signal of interest adaptively which retrain the energy in signal subspace and neglect the energy in noise subspace. Here, we develop an adaptive rank estimation by the SVD for denoising purpose based on estimating the noise level σest using the first level detail symmlet-wavelet's coefficients d1. The optimal rank is obtained at the point where the difference between the noisy and the reduced rank dataset is approximately below the estimated noise level. The proposed method has successfully estimated the optimal rank which gives the best denoising performance.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130820296","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677887
Sagila Gangadharan K, Benzy V. K, A. Vinod
Motor-Imagery-based Brain Computer Interface (MI-BCI) decodes the parameters of imagined motor movement and translates it into control commands to the external world. It has potential applications in neurorehabilitation and development of assistive technology. This paper investigates the Electroencephalogram (EEG) correlates of direction parameters of a center-out hand movement imagination task in right and left directions. A variance-based time bin selection algorithm is proposed to select the most discriminative EEG time segment for directional classification of movement imagination. The discriminative EEG features carrying motor imagery (MI) directional information are extracted from the selected EEG time segment using the wavelet-common spatial pattern (WCSP) algorithm. The WCSP features are classified using Support Vector Machine classifier resulting in a cross validated classification accuracy of 71% between left versus right MI directions of 15 subjects.
{"title":"Maximum Variance-based EEG Time Bin Selection for Decoding of Imagined Hand Movement Directions in Brain Computer Interface","authors":"Sagila Gangadharan K, Benzy V. K, A. Vinod","doi":"10.1109/BioSMART54244.2021.9677887","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677887","url":null,"abstract":"Motor-Imagery-based Brain Computer Interface (MI-BCI) decodes the parameters of imagined motor movement and translates it into control commands to the external world. It has potential applications in neurorehabilitation and development of assistive technology. This paper investigates the Electroencephalogram (EEG) correlates of direction parameters of a center-out hand movement imagination task in right and left directions. A variance-based time bin selection algorithm is proposed to select the most discriminative EEG time segment for directional classification of movement imagination. The discriminative EEG features carrying motor imagery (MI) directional information are extracted from the selected EEG time segment using the wavelet-common spatial pattern (WCSP) algorithm. The WCSP features are classified using Support Vector Machine classifier resulting in a cross validated classification accuracy of 71% between left versus right MI directions of 15 subjects.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134380830","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677858
Navid Shaahaghi, Supriya Karishetti, Nancy Ma
Influenza, or most commonly termed the flu, is a common respiratory illness caused by viral infection. The circulation of this virus is found year-round but is more common during the flu season: fall and winter. In the United States, the number of reported cases begins to rise in October, reaches a peak in December, and returns to normal in April. Even though there are four subtypes of the Influenza virus, the seasonal flu outbreaks in humans are caused by type A and B viruses. eVision utilizes influenza data provided by the United States Center for Disease Control and Prevention (CDC) and the World Health Organization (WHO) to analyze influenza A and B cases throughout the flu season. During the 2019–20 flu season, the positive influenza cases reported in the US were between 36 and 56 million, which is the highest over the past six years. However, during the 2020–21 flu season which is the first complete flu season within the COVID-19 pandemic, the reported flu cases reduced drastically to 1,899; of which 713 were caused by influenza A viruses, and 1,186 by influenza B viruses. This indicates that the number of flu B cases was higher than that of flu A which was not normally the case prior to the COVID-19 pandemic. It was further observed that flu B reached its peak either at the same time or earlier than flu A which is also unusual compared to the flu trends prior to the onset of the COVID-19 pandemic. This peculiar trend is also noted during the Severe Acute Respiratory Syndrome (SARS) outbreak in 2003. This paper reports the findings on deviation in the Influenza type A and type B trends during the circulation of Coronavirus in the US and Canada and provides possible reasons for these changes.
{"title":"Interplay of Influenza A/B Subtypes and COVID-19","authors":"Navid Shaahaghi, Supriya Karishetti, Nancy Ma","doi":"10.1109/BioSMART54244.2021.9677858","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677858","url":null,"abstract":"Influenza, or most commonly termed the flu, is a common respiratory illness caused by viral infection. The circulation of this virus is found year-round but is more common during the flu season: fall and winter. In the United States, the number of reported cases begins to rise in October, reaches a peak in December, and returns to normal in April. Even though there are four subtypes of the Influenza virus, the seasonal flu outbreaks in humans are caused by type A and B viruses. eVision utilizes influenza data provided by the United States Center for Disease Control and Prevention (CDC) and the World Health Organization (WHO) to analyze influenza A and B cases throughout the flu season. During the 2019–20 flu season, the positive influenza cases reported in the US were between 36 and 56 million, which is the highest over the past six years. However, during the 2020–21 flu season which is the first complete flu season within the COVID-19 pandemic, the reported flu cases reduced drastically to 1,899; of which 713 were caused by influenza A viruses, and 1,186 by influenza B viruses. This indicates that the number of flu B cases was higher than that of flu A which was not normally the case prior to the COVID-19 pandemic. It was further observed that flu B reached its peak either at the same time or earlier than flu A which is also unusual compared to the flu trends prior to the onset of the COVID-19 pandemic. This peculiar trend is also noted during the Severe Acute Respiratory Syndrome (SARS) outbreak in 2003. This paper reports the findings on deviation in the Influenza type A and type B trends during the circulation of Coronavirus in the US and Canada and provides possible reasons for these changes.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129730977","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 : 2021-12-08DOI: 10.1109/BioSMART54244.2021.9677572
Hazem Zein, S. Chantaf, Rola El-Saleh, A. Nait-Ali
Deep-Learning based approaches in dermatology face a significant problem regarding the availability of free open datasets. Recently, Generative Adversarial Networks (GANs) were successfully employed to generate artificial images through a combination of two models: Generator and Discriminator. In this work, we propose using StyleGAN2 to generate realistic artificial faces presenting acne diseases. The model uses a collection of authentic face images gathered from multiple sources regardless of the acquisition conditions such as resolution, pose, Etc. Results show that the model can produce an unlimited number of artificial faces of acne diseases. The biomedical community can take advantage of such a dataset to evaluate the performance of some specific algorithms.
{"title":"Generative Adversarial Networks Based Approach for Artificial Face Dataset Generation in Acne Disease Cases","authors":"Hazem Zein, S. Chantaf, Rola El-Saleh, A. Nait-Ali","doi":"10.1109/BioSMART54244.2021.9677572","DOIUrl":"https://doi.org/10.1109/BioSMART54244.2021.9677572","url":null,"abstract":"Deep-Learning based approaches in dermatology face a significant problem regarding the availability of free open datasets. Recently, Generative Adversarial Networks (GANs) were successfully employed to generate artificial images through a combination of two models: Generator and Discriminator. In this work, we propose using StyleGAN2 to generate realistic artificial faces presenting acne diseases. The model uses a collection of authentic face images gathered from multiple sources regardless of the acquisition conditions such as resolution, pose, Etc. Results show that the model can produce an unlimited number of artificial faces of acne diseases. The biomedical community can take advantage of such a dataset to evaluate the performance of some specific algorithms.","PeriodicalId":286026,"journal":{"name":"2021 4th International Conference on Bio-Engineering for Smart Technologies (BioSMART)","volume":"127 27","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133085683","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：