Pub Date : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079609
Y. Z. Chong, C. Tan, S. Chan, Choon-Hian Goh
There are various upper-extremities rehabilitation systems available to cater for various communities-in-need. Majority of these systems have only a single pre-programmed protocol. This paper aims to design and develop an affordable (~MYR 230), light-weight (~250 g) exoskeleton system where the transmission mechanism is based on elements of origami string technique. The features of the system include light-mass; passive movement rehabilitation of the finger (flexion-extension at MCP and PIP joints) and wrist (extension at radiocarpal joint); user-centric rehabilitation protocols; status review of rehabilitation. The system also incorporated sensing systems (flex sensors) to record various biomechanical measurements; control system (ESP32); actuator system (MG995 servo motors) alongside with a mobile application that allows the user to select the rehabilitation protocols, view rehabilitation progress and learn how to fold origami models. It is anticipated that this development will be a platform to further explore on the adaptation of origami techniques in development of rehabilitation devices.
{"title":"Development of IoT-Based, Origami-Inspired Wearable Rehabilitation Device for Wrist-Finger Mobility Rehabilitation","authors":"Y. Z. Chong, C. Tan, S. Chan, Choon-Hian Goh","doi":"10.1109/IECBES54088.2022.10079609","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079609","url":null,"abstract":"There are various upper-extremities rehabilitation systems available to cater for various communities-in-need. Majority of these systems have only a single pre-programmed protocol. This paper aims to design and develop an affordable (~MYR 230), light-weight (~250 g) exoskeleton system where the transmission mechanism is based on elements of origami string technique. The features of the system include light-mass; passive movement rehabilitation of the finger (flexion-extension at MCP and PIP joints) and wrist (extension at radiocarpal joint); user-centric rehabilitation protocols; status review of rehabilitation. The system also incorporated sensing systems (flex sensors) to record various biomechanical measurements; control system (ESP32); actuator system (MG995 servo motors) alongside with a mobile application that allows the user to select the rehabilitation protocols, view rehabilitation progress and learn how to fold origami models. It is anticipated that this development will be a platform to further explore on the adaptation of origami techniques in development of rehabilitation devices.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124258049","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079652
Christopher Yew Shuen Ang, N. L. Loo, Y. Chiew, C. P. Tan, M. Nor, J. Chase
Asynchronous breathing (AB) in mechanical ventilation (MV) patients is heterogenous, patient-specific, and is associated with adverse patient outcomes. Various machine learning models have been developed for AB detection, however studies regarding the data structures used for model training are scarce. This study investigates the effects of different training data structures and sizes of Convolutional Neural Networks (CNN) to detect AB. Four CNN models were developed using different amounts of data and data structures: one-dimension, line, area, and array. Training datasets consisting of 300, 1,000, 5,000 and 10,000 airway pressure waveforms from MV patients were used for model development. Model sensitivity and specificity were evaluated using an independent set of 3000 waveforms in a 100-iteration Monte-Carlo analysis. The best-performing CNN model was used to determine Asynchrony Index (AI) values in a clinical patient cohort. Monte-Carlo analysis showed that models trained with datasets of 10,000 breathing cycles delivered $gt99$% sensitivity and specificity. Relatively lower sensitivity and specificity of $lt78.8$% and $lt96.7$% respectively were obtained when trained with data quantities of 5000 breaths or less. A CNN trained with 1Dimensional data structure yielded 99.9% sensitivity and 99.6% specificity. It achieved 88.5% average accuracy when validated with an independent clinical data set of 544,319 breaths. Asynchrony breathing detection is ubiquitous and 1-Dimensional data structures provide a resource efficient method for the development of an accurate CNN model.
{"title":"Effects of Data Structure in Convolutional Neural Network for Detection of Asynchronous Breathing in Mechanical Ventilation Treatment","authors":"Christopher Yew Shuen Ang, N. L. Loo, Y. Chiew, C. P. Tan, M. Nor, J. Chase","doi":"10.1109/IECBES54088.2022.10079652","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079652","url":null,"abstract":"Asynchronous breathing (AB) in mechanical ventilation (MV) patients is heterogenous, patient-specific, and is associated with adverse patient outcomes. Various machine learning models have been developed for AB detection, however studies regarding the data structures used for model training are scarce. This study investigates the effects of different training data structures and sizes of Convolutional Neural Networks (CNN) to detect AB. Four CNN models were developed using different amounts of data and data structures: one-dimension, line, area, and array. Training datasets consisting of 300, 1,000, 5,000 and 10,000 airway pressure waveforms from MV patients were used for model development. Model sensitivity and specificity were evaluated using an independent set of 3000 waveforms in a 100-iteration Monte-Carlo analysis. The best-performing CNN model was used to determine Asynchrony Index (AI) values in a clinical patient cohort. Monte-Carlo analysis showed that models trained with datasets of 10,000 breathing cycles delivered $gt99$% sensitivity and specificity. Relatively lower sensitivity and specificity of $lt78.8$% and $lt96.7$% respectively were obtained when trained with data quantities of 5000 breaths or less. A CNN trained with 1Dimensional data structure yielded 99.9% sensitivity and 99.6% specificity. It achieved 88.5% average accuracy when validated with an independent clinical data set of 544,319 breaths. Asynchrony breathing detection is ubiquitous and 1-Dimensional data structures provide a resource efficient method for the development of an accurate CNN model.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122926909","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079445
Xin Yang, R. Rimal, Tiffany Rogers
Due to the continuous advances in brain imaging technology, an increasing number of large-scale brain research projects have been derived, such as the ABIDE Initiative. These developments have enabled us to gain an unprecedentedly detailed insight into brain activity by analyzing brain imaging data, which will reshape our understanding of brain activity and uncover biomarkers of brain disease.Over the past decade, the analysis of resting-state functional connectivity has become a trend because brain connectivity provides an effective way to understand how spatially distant brain regions interact and achieve coherent neural functions. One of the most common approaches to analyze functional connectivity is the Pearson correlation. This paper uses a new correlation method to calculate functional connectivity: Spearman’s rank correlation. We apply two conventional machine learning methods to classify autism spectrum disorder (ASD) patients from typically developing (TD) participants based on functional connectivity derived from resting-state functional magnetic resonance imaging (fMRI) data. To verify the feasibility and validity of Spearman’s rank correlation in the classification of autism, we compared the accuracy, sensitivity, and specificity of methods using functional connectivity obtained from Pearson’s correlation and Spearman’s rank correlation. Moreover, feature selection is one of the essential tasks in classification studies. We present an empirical comparison of two feature selection methods: select from model (SFM) and recursive feature elimination (RFE).
{"title":"Functional Connectivity Based Classification for Autism Spectrum Disorder Using Spearman’s Rank Correlation","authors":"Xin Yang, R. Rimal, Tiffany Rogers","doi":"10.1109/IECBES54088.2022.10079445","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079445","url":null,"abstract":"Due to the continuous advances in brain imaging technology, an increasing number of large-scale brain research projects have been derived, such as the ABIDE Initiative. These developments have enabled us to gain an unprecedentedly detailed insight into brain activity by analyzing brain imaging data, which will reshape our understanding of brain activity and uncover biomarkers of brain disease.Over the past decade, the analysis of resting-state functional connectivity has become a trend because brain connectivity provides an effective way to understand how spatially distant brain regions interact and achieve coherent neural functions. One of the most common approaches to analyze functional connectivity is the Pearson correlation. This paper uses a new correlation method to calculate functional connectivity: Spearman’s rank correlation. We apply two conventional machine learning methods to classify autism spectrum disorder (ASD) patients from typically developing (TD) participants based on functional connectivity derived from resting-state functional magnetic resonance imaging (fMRI) data. To verify the feasibility and validity of Spearman’s rank correlation in the classification of autism, we compared the accuracy, sensitivity, and specificity of methods using functional connectivity obtained from Pearson’s correlation and Spearman’s rank correlation. Moreover, feature selection is one of the essential tasks in classification studies. We present an empirical comparison of two feature selection methods: select from model (SFM) and recursive feature elimination (RFE).","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114493463","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079260
F. Plocksties, A. Lüttig, Christoph Niemann, Felix Uster, D. Franz, Maria Kober, Maximilian Koschay, S. Perl, A. Richter, R. Köhling, Alexander Storch, D. Timmermann
Deep brain stimulation (DBS) is an essential therapeutic resource for treating movement disorders like dystonia. In order to gain further insight into the underlying disease mechanisms, animal models are used. However, the most critical obstacle for further research is the lack of subcutaneous implantable, miniaturized neurostimulators that can deliver reliable and replicable results. Extracorporeal mounting of neurostimulators on the head or back places a high burden on the animal. Furthermore, the animals frequently tamper with these stimulation setups, leading to high failure rates. In the absence of suitable diagnostic tests, such defects generally escape detection. Therefore, this article presents strategies for a DBS stimulator design intending to increase the scientific merit of behavioral experiments in rodents. In this context, we demonstrate an easy-to-implement, waterproof, biocompatible, and compact encapsulation method suited for full implantation in small rodents. Using this method, we implanted DBS devices subcutaneously in dystonic hamsters that have been successfully tested for up to 17 days.
{"title":"Strategies on Deep Brain Stimulation Devices for Effective Behavioral Studies in Rodents","authors":"F. Plocksties, A. Lüttig, Christoph Niemann, Felix Uster, D. Franz, Maria Kober, Maximilian Koschay, S. Perl, A. Richter, R. Köhling, Alexander Storch, D. Timmermann","doi":"10.1109/IECBES54088.2022.10079260","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079260","url":null,"abstract":"Deep brain stimulation (DBS) is an essential therapeutic resource for treating movement disorders like dystonia. In order to gain further insight into the underlying disease mechanisms, animal models are used. However, the most critical obstacle for further research is the lack of subcutaneous implantable, miniaturized neurostimulators that can deliver reliable and replicable results. Extracorporeal mounting of neurostimulators on the head or back places a high burden on the animal. Furthermore, the animals frequently tamper with these stimulation setups, leading to high failure rates. In the absence of suitable diagnostic tests, such defects generally escape detection. Therefore, this article presents strategies for a DBS stimulator design intending to increase the scientific merit of behavioral experiments in rodents. In this context, we demonstrate an easy-to-implement, waterproof, biocompatible, and compact encapsulation method suited for full implantation in small rodents. Using this method, we implanted DBS devices subcutaneously in dystonic hamsters that have been successfully tested for up to 17 days.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123946649","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}
During the COVID-19 outbreak, many healthcare workers (HCWs) have been infected because they failed to comply with the correct process of donning and doffing personal protective equipment (PPE). Based on this, we develop a gesture-controlled system that not only can train HCWs but also can give HCWs real-time guidance during the process of donning and doffing PPE. It can effectively prevent the infection of HCWs. We first use the hand detection algorithm to locate the position of the HCWs, helping them to enter the proper area. Then they can use our gesture recognition algorithm to control the playback of the videos which guides them in donning and doffing PPE. We verify the effectiveness of the system through a series of experiments. The results show the great value of our system in the protection of HCWs.
{"title":"A Gesture Controlled System to Train and Guide the Healthcare Workers While Donning and Doffing Personal Protective Equipment","authors":"Penglin Qin, Guanghua Xu, Qingqiang Wu, Fan Wei, Zejiang Li, Dakai Zhao","doi":"10.1109/IECBES54088.2022.10079337","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079337","url":null,"abstract":"During the COVID-19 outbreak, many healthcare workers (HCWs) have been infected because they failed to comply with the correct process of donning and doffing personal protective equipment (PPE). Based on this, we develop a gesture-controlled system that not only can train HCWs but also can give HCWs real-time guidance during the process of donning and doffing PPE. It can effectively prevent the infection of HCWs. We first use the hand detection algorithm to locate the position of the HCWs, helping them to enter the proper area. Then they can use our gesture recognition algorithm to control the playback of the videos which guides them in donning and doffing PPE. We verify the effectiveness of the system through a series of experiments. The results show the great value of our system in the protection of HCWs.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121511745","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079657
Krishnaveni Parvataneni, S. Zaidi
Due to their low gas temperatures, non-thermal, dielectric barrier discharge (DBD) finds frequent applications in wound healing and sterilization. Reactive nitrogen and oxygen species in the plasma play a vital role in this process. The concentration of these radicals is dependent on the plasma’s operating conditions (e.g. applied voltage and gas flow rates). Radicals’ vibrational and rotational temperatures play a vital role in wound healing and are required to find an optimized plasma operating condition for the wound healing process. The current project aims to measure the plasma temperatures (vibrational, rotational, and electronic) for a multi-electrode torch. The emission spectrum for various electrodes of the plasma torch was captured at various operating conditions. For this purpose, an Ocean Optics UV-IR spectrometer in conjunction with SPECAIR was used to estimate the vibrational, rotational, and electronic temperatures of the plasma. For this experiment the multi-electrode plasma torch was operated at various conditions by changing the outer electrodes, gas flow rates (helium 20-40 slpm), and input applied voltages (5kV-10kV and 20-40 kHZ). Experimental results reveal that both plasma vibrational and rotational temperatures $(sim 500mathrm{K}700mathrm{K})$ were dependent on the measuring position from the plasma exit at identical operating conditions. No significant change in electronic temperatures $(sim 2800mathrm{K}$ – 3000K) was observed for all conditions. Detailed results are included in this paper.
{"title":"Exploring the Vibrational and Rotational Temperatures of a DBD Plasma Jet for Wound Healing","authors":"Krishnaveni Parvataneni, S. Zaidi","doi":"10.1109/IECBES54088.2022.10079657","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079657","url":null,"abstract":"Due to their low gas temperatures, non-thermal, dielectric barrier discharge (DBD) finds frequent applications in wound healing and sterilization. Reactive nitrogen and oxygen species in the plasma play a vital role in this process. The concentration of these radicals is dependent on the plasma’s operating conditions (e.g. applied voltage and gas flow rates). Radicals’ vibrational and rotational temperatures play a vital role in wound healing and are required to find an optimized plasma operating condition for the wound healing process. The current project aims to measure the plasma temperatures (vibrational, rotational, and electronic) for a multi-electrode torch. The emission spectrum for various electrodes of the plasma torch was captured at various operating conditions. For this purpose, an Ocean Optics UV-IR spectrometer in conjunction with SPECAIR was used to estimate the vibrational, rotational, and electronic temperatures of the plasma. For this experiment the multi-electrode plasma torch was operated at various conditions by changing the outer electrodes, gas flow rates (helium 20-40 slpm), and input applied voltages (5kV-10kV and 20-40 kHZ). Experimental results reveal that both plasma vibrational and rotational temperatures $(sim 500mathrm{K}700mathrm{K})$ were dependent on the measuring position from the plasma exit at identical operating conditions. No significant change in electronic temperatures $(sim 2800mathrm{K}$ – 3000K) was observed for all conditions. Detailed results are included in this paper.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127925115","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079453
M. S. Abdullah, A. Radzol, M. Marzuki, K. Y. Lee, S. A. Ahmad
Coronavirus disease (COVID-19) is an infectious disease caused by the coronavirus was first found in Wuhan, China in December 2019. It has infected more than 300 million people with more than 5 million of death cases. Until now, the virus is still evolving producing new variants of concern contributes to the increase the infection rate around the world. Thus, various diagnostic procedures are in need to help physicians in diagnosis disease certainly and rapidly. In this study, deep learning approach is used to classify normal and COVID-19 cases from CT scan images. Normalizer Free CNN network (NFNets) model is implemented on the images. Statistical measures such as accuracy, precision, sensitivity (also known as recall) are used to evaluate the performance of the model against the previous studies. Loss of 0.0842, accuracy of 0.7227, precision of 0.9751 and recall of 0.9727 are achieved. Thus, further optimization on the NFNets learning algorithm is required to improve the classification performanceClinical Relevance–Implementation of deep learning technique to automate diagnosis of diseases such as COVID-19 cases from CT scan images will simplify the clinical flow towards providing reliable intelligent aids for patient care.
{"title":"NFNets-CNN for Classification of COVID-19 from CT Scan Images","authors":"M. S. Abdullah, A. Radzol, M. Marzuki, K. Y. Lee, S. A. Ahmad","doi":"10.1109/IECBES54088.2022.10079453","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079453","url":null,"abstract":"Coronavirus disease (COVID-19) is an infectious disease caused by the coronavirus was first found in Wuhan, China in December 2019. It has infected more than 300 million people with more than 5 million of death cases. Until now, the virus is still evolving producing new variants of concern contributes to the increase the infection rate around the world. Thus, various diagnostic procedures are in need to help physicians in diagnosis disease certainly and rapidly. In this study, deep learning approach is used to classify normal and COVID-19 cases from CT scan images. Normalizer Free CNN network (NFNets) model is implemented on the images. Statistical measures such as accuracy, precision, sensitivity (also known as recall) are used to evaluate the performance of the model against the previous studies. Loss of 0.0842, accuracy of 0.7227, precision of 0.9751 and recall of 0.9727 are achieved. Thus, further optimization on the NFNets learning algorithm is required to improve the classification performanceClinical Relevance–Implementation of deep learning technique to automate diagnosis of diseases such as COVID-19 cases from CT scan images will simplify the clinical flow towards providing reliable intelligent aids for patient care.","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132607008","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 : 2022-12-07DOI: 10.1109/IECBES54088.2022.10079331
Der Sheng Tan, Wei Qiang Tam, H. Nisar, K. Yeap
To aid in the clinical diagnosis of brain tumors, magnetic resonance imaging (MRI) is frequently used. The amount of time it takes to manually segment MRI images depends on the radiologist’s level of expertise. In this paper, a novel U-Net architecture for segmenting images of brain tumors is proposed. We have evaluated BraTS 2020 dataset with an improved U-Net structure with a dropout layer inserted between the encoder and decoder to reduce overfitting. By comparing with other U-Net architectures, our method has shown a promising result with dice coefficients 70.40%, 69.08% and 73.03%, for whole tumor (WT), tumor core (TC) and enhanced tumor (ET).
{"title":"Segmenting Brain Tumor with an Improved U-Net Architecture","authors":"Der Sheng Tan, Wei Qiang Tam, H. Nisar, K. Yeap","doi":"10.1109/IECBES54088.2022.10079331","DOIUrl":"https://doi.org/10.1109/IECBES54088.2022.10079331","url":null,"abstract":"To aid in the clinical diagnosis of brain tumors, magnetic resonance imaging (MRI) is frequently used. The amount of time it takes to manually segment MRI images depends on the radiologist’s level of expertise. In this paper, a novel U-Net architecture for segmenting images of brain tumors is proposed. We have evaluated BraTS 2020 dataset with an improved U-Net structure with a dropout layer inserted between the encoder and decoder to reduce overfitting. By comparing with other U-Net architectures, our method has shown a promising result with dice coefficients 70.40%, 69.08% and 73.03%, for whole tumor (WT), tumor core (TC) and enhanced tumor (ET).","PeriodicalId":146681,"journal":{"name":"2022 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130743262","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：