This study's purpose is to assess the stress distribution in the peri-implant bone, implants, and prosthetic framework using two different posterior implant angles. All-on-four maxillary prostheses fabricated from feldspathic-ceramic-veneered zirconia-reinforced lithium silicate (ZLS) and feldspathic-ceramic-veneered cobalt-chromium (CoCr) were designed with 17 or 30-degree-angled posterior implants. Posterior cantilever and frontal vertical loads were applied to all models. The distribution of maximum and minimum principal stresses (σmax and σmin) and von Mises stress (σVM) was evaluated. Under posterior cantilever load, with an increase in posterior implant angle, σmax decreased by 4 and 7 MPa in the cortical bone when ZLS and CoCr were used as a prosthetic framework, respectively. Regardless of the framework material, 17-degree-angled posterior implants showed the highest σVM (541.36 MPa under posterior cantilever load; 110.79 MPa under frontal vertical load) values. Regardless of the posterior implant angle, ZLS framework showed the highest σVM (91.59 MPa under posterior cantilever load; 218.99 MPa under frontal vertical load) values. Increasing implant angle from 17 to 30° caused a decrease in σmax values in the cortical bone. Designs with 30-degree posterior implant angles and ZLS framework material may be preferred in All-on-four implant-supported fixed complete dentures.
{"title":"Mechanical response of different frameworks for maxillary all-on-four implant-supported fixed dental prosthesis: 3D finite element analysis.","authors":"Zekiye Begüm Güçlü, Ayhan Gürbüz, Gonca Deste Gökay, Rukiye Durkan, Perihan Oyar","doi":"10.1515/bmt-2022-0176","DOIUrl":"https://doi.org/10.1515/bmt-2022-0176","url":null,"abstract":"<p><p>This study's purpose is to assess the stress distribution in the peri-implant bone, implants, and prosthetic framework using two different posterior implant angles. All-on-four maxillary prostheses fabricated from feldspathic-ceramic-veneered zirconia-reinforced lithium silicate (ZLS) and feldspathic-ceramic-veneered cobalt-chromium (CoCr) were designed with 17 or 30-degree-angled posterior implants. Posterior cantilever and frontal vertical loads were applied to all models. The distribution of maximum and minimum principal stresses (σmax and σmin) and von Mises stress (σVM) was evaluated. Under posterior cantilever load, with an increase in posterior implant angle, σmax decreased by 4 and 7 MPa in the cortical bone when ZLS and CoCr were used as a prosthetic framework, respectively. Regardless of the framework material, 17-degree-angled posterior implants showed the highest σVM (541.36 MPa under posterior cantilever load; 110.79 MPa under frontal vertical load) values. Regardless of the posterior implant angle, ZLS framework showed the highest σVM (91.59 MPa under posterior cantilever load; 218.99 MPa under frontal vertical load) values. Increasing implant angle from 17 to 30° caused a decrease in σmax values in the cortical bone. Designs with 30-degree posterior implant angles and ZLS framework material may be preferred in All-on-four implant-supported fixed complete dentures.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40617677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-15Print Date: 2022-10-26DOI: 10.1515/bmt-2021-0019
Geoffrey Z Iwata, Yinan Hu, Arne Wickenbrock, Tilmann Sander, Muthuraman Muthuraman, Venkata Chaitanya Chirumamilla, Sergiu Groppa, Qishan Liu, Dmitry Budker
Transcranial magnetic stimulation (TMS) has widespread clinical applications from diagnosis to treatment. We combined TMS with non-contact magnetic detection of TMS-evoked muscle activity in peripheral limbs to explore a new diagnostic modality that enhances the utility of TMS as a clinical tool by leveraging technological advances in magnetometry. We recorded measurements in a regular hospital room using an array of optically pumped magnetometers (OPMs) inside a portable shield that encloses only the forearm and hand of the subject. We present magnetomyograms (MMG)s of TMS-evoked movement in a human hand, together with a simultaneous surface electromyograph (EMG) data. The biomagnetic signals recorded in the MMG provides detailed spatial and temporal information that is complementary to that of the electric signal channels. Moreover, we identify features in the magnetic recording beyond that of the EMG. This system demonstrates the value of biomagnetic signals in TMS-based clinical approaches and widens its availability and practical potential.
{"title":"Biomagnetic signals recorded during transcranial magnetic stimulation (TMS)-evoked peripheral muscular activity.","authors":"Geoffrey Z Iwata, Yinan Hu, Arne Wickenbrock, Tilmann Sander, Muthuraman Muthuraman, Venkata Chaitanya Chirumamilla, Sergiu Groppa, Qishan Liu, Dmitry Budker","doi":"10.1515/bmt-2021-0019","DOIUrl":"https://doi.org/10.1515/bmt-2021-0019","url":null,"abstract":"<p><p>Transcranial magnetic stimulation (TMS) has widespread clinical applications from diagnosis to treatment. We combined TMS with non-contact magnetic detection of TMS-evoked muscle activity in peripheral limbs to explore a new diagnostic modality that enhances the utility of TMS as a clinical tool by leveraging technological advances in magnetometry. We recorded measurements in a regular hospital room using an array of optically pumped magnetometers (OPMs) inside a portable shield that encloses only the forearm and hand of the subject. We present magnetomyograms (MMG)s of TMS-evoked movement in a human hand, together with a simultaneous surface electromyograph (EMG) data. The biomagnetic signals recorded in the MMG provides detailed spatial and temporal information that is complementary to that of the electric signal channels. Moreover, we identify features in the magnetic recording beyond that of the EMG. This system demonstrates the value of biomagnetic signals in TMS-based clinical approaches and widens its availability and practical potential.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40609229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-08Print Date: 2022-10-26DOI: 10.1515/bmt-2022-0180
Zakaria Neili, Kenneth Sundaraj
In lung sound classification using deep learning, many studies have considered the use of short-time Fourier transform (STFT) as the most commonly used 2D representation of the input data. Consequently, STFT has been widely used as an analytical tool, but other versions of the representation have also been developed. This study aims to evaluate and compare the performance of the spectrogram, scalogram, melspectrogram and gammatonegram representations, and provide comparative information to users regarding the suitability of these time-frequency (TF) techniques in lung sound classification. Lung sound signals used in this study were obtained from the ICBHI 2017 respiratory sound database. These lung sound recordings were converted into images of spectrogram, scalogram, melspectrogram and gammatonegram TF representations respectively. The four types of images were fed separately into the VGG16, ResNet-50 and AlexNet deep-learning architectures. Network performances were analyzed and compared based on accuracy, precision, recall and F1-score. The results of the analysis on the performance of the four representations using these three commonly used CNN deep-learning networks indicate that the generated gammatonegram and scalogram TF images coupled with ResNet-50 achieved maximum classification accuracies.
{"title":"A comparative study of the spectrogram, scalogram, melspectrogram and gammatonegram time-frequency representations for the classification of lung sounds using the ICBHI database based on CNNs.","authors":"Zakaria Neili, Kenneth Sundaraj","doi":"10.1515/bmt-2022-0180","DOIUrl":"https://doi.org/10.1515/bmt-2022-0180","url":null,"abstract":"<p><p>In lung sound classification using deep learning, many studies have considered the use of short-time Fourier transform (STFT) as the most commonly used 2D representation of the input data. Consequently, STFT has been widely used as an analytical tool, but other versions of the representation have also been developed. This study aims to evaluate and compare the performance of the spectrogram, scalogram, melspectrogram and gammatonegram representations, and provide comparative information to users regarding the suitability of these time-frequency (TF) techniques in lung sound classification. Lung sound signals used in this study were obtained from the ICBHI 2017 respiratory sound database. These lung sound recordings were converted into images of spectrogram, scalogram, melspectrogram and gammatonegram TF representations respectively. The four types of images were fed separately into the VGG16, ResNet-50 and AlexNet deep-learning architectures. Network performances were analyzed and compared based on accuracy, precision, recall and F1-score. The results of the analysis on the performance of the four representations using these three commonly used CNN deep-learning networks indicate that the generated gammatonegram and scalogram TF images coupled with ResNet-50 achieved maximum classification accuracies.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40601433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-04Print Date: 2022-10-26DOI: 10.1515/bmt-2022-0067
Tao Wang, Changhua Lu, Yining Sun, Hengyang Fang, Weiwei Jiang, Chun Liu
Sleep apnea is a sleep disorder caused by weakened or suspended breathing during sleep, which seriously affects the work and health of patients. The traditional polysomnography (PSG) detection process is complicated and expensive, which has attracted researchers to explore a rapid detection method based on single-lead ECG signals. However, existing ECG-based sleep apnea detection methods have certain limitations and complexities, mainly relying on human-crafted features. To solve the problem, the paper develops a sleep apnea detection method based on a residual attention mechanism network. The method uses the RR interval signal and the R-peak signal derived from the ECG signal as input, realizes feature extraction through the residual network (ResNet), and adds the SENet attention mechanism to deepen the mining of channel features. Experimental results show that the per-segment accuracy of the proposed method can reach 86.2%. Compared with existing works, its accuracy has increased by 1.1-8.1%. These results show that the proposed residual attention network can effectively use ECG signals to quickly detect sleep apnea. Meanwhile, compared with existing works, the proposed method overcomes the limitations and complexity of human-crafted features in sleep apnea detection research.
{"title":"A method to detect sleep apnea using residual attention mechanism network from single-lead ECG signal.","authors":"Tao Wang, Changhua Lu, Yining Sun, Hengyang Fang, Weiwei Jiang, Chun Liu","doi":"10.1515/bmt-2022-0067","DOIUrl":"https://doi.org/10.1515/bmt-2022-0067","url":null,"abstract":"<p><p>Sleep apnea is a sleep disorder caused by weakened or suspended breathing during sleep, which seriously affects the work and health of patients. The traditional polysomnography (PSG) detection process is complicated and expensive, which has attracted researchers to explore a rapid detection method based on single-lead ECG signals. However, existing ECG-based sleep apnea detection methods have certain limitations and complexities, mainly relying on human-crafted features. To solve the problem, the paper develops a sleep apnea detection method based on a residual attention mechanism network. The method uses the RR interval signal and the R-peak signal derived from the ECG signal as input, realizes feature extraction through the residual network (ResNet), and adds the SENet attention mechanism to deepen the mining of channel features. Experimental results show that the per-segment accuracy of the proposed method can reach 86.2%. Compared with existing works, its accuracy has increased by 1.1-8.1%. These results show that the proposed residual attention network can effectively use ECG signals to quickly detect sleep apnea. Meanwhile, compared with existing works, the proposed method overcomes the limitations and complexity of human-crafted features in sleep apnea detection research.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40667135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Abstracts of the 2022 Joint Annual Conference of the Austrian (ÖGBMT), German (VDE DGBMT) and Swiss (SSBE) Societies for Biomedical Engineering, including the 14th Vienna International Workshop on Functional Electrical Stimulation","authors":"D. Baumgarten","doi":"10.1515/bmt-2022-2001","DOIUrl":"https://doi.org/10.1515/bmt-2022-2001","url":null,"abstract":"","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87001561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-15Print Date: 2022-10-26DOI: 10.1515/bmt-2021-0044
Smitha Balakrishnan, Paul K Joseph
Myocardial infarction is one of the major life-threatening diseases. The cause is atherosclerosis i.e. the occlusion of the coronary artery by deposition of plaque on its walls. The severity of plaque deposition in the artery depends on the characteristics of the plaque. Hence, the classification of the type of plaque is crucial for assessing the risk of atherosclerosis and predicting the chances of myocardial infarction. This paper proposes prediction of atherosclerotic risk by non-invasive ultrasound image segmentation and textural feature extraction. The intima-media complex is segmented using a snakes-based segmentation algorithm on the arterial wall in the ultrasound images. Then, the plaque is extracted from the segmented intima-media complex. The features of the plaque are obtained by computing Hu's moment invariants. Visual pattern recognition independent of position, size, orientation and parallel projection could be done using these moment invariants. For the classification of the features of the plaque, an SVM classifier is used. The performance shows improvement in accuracy using lesser number of features than previous works. The reduction in feature size is achieved by incorporating segmentation in the pre-processing stage. Tenfold cross-validation protocol is used for training and testing the classifier. An accuracy of 97.9% is obtained with only two features. This proposed technique could work as an adjunct tool in quick decision-making for cardiologists and radiologists. The segmentation step introduced in the preprocessing stage improved the feature extraction technique. An improvement in performance is achieved with much less number of features.
{"title":"Stratification of risk of atherosclerotic plaque using Hu's moment invariants of segmented ultrasonic images.","authors":"Smitha Balakrishnan, Paul K Joseph","doi":"10.1515/bmt-2021-0044","DOIUrl":"https://doi.org/10.1515/bmt-2021-0044","url":null,"abstract":"<p><p>Myocardial infarction is one of the major life-threatening diseases. The cause is atherosclerosis i.e. the occlusion of the coronary artery by deposition of plaque on its walls. The severity of plaque deposition in the artery depends on the characteristics of the plaque. Hence, the classification of the type of plaque is crucial for assessing the risk of atherosclerosis and predicting the chances of myocardial infarction. This paper proposes prediction of atherosclerotic risk by non-invasive ultrasound image segmentation and textural feature extraction. The intima-media complex is segmented using a snakes-based segmentation algorithm on the arterial wall in the ultrasound images. Then, the plaque is extracted from the segmented intima-media complex. The features of the plaque are obtained by computing Hu's moment invariants. Visual pattern recognition independent of position, size, orientation and parallel projection could be done using these moment invariants. For the classification of the features of the plaque, an SVM classifier is used. The performance shows improvement in accuracy using lesser number of features than previous works. The reduction in feature size is achieved by incorporating segmentation in the pre-processing stage. Tenfold cross-validation protocol is used for training and testing the classifier. An accuracy of 97.9% is obtained with only two features. This proposed technique could work as an adjunct tool in quick decision-making for cardiologists and radiologists. The segmentation step introduced in the preprocessing stage improved the feature extraction technique. An improvement in performance is achieved with much less number of features.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40534458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-04Print Date: 2022-10-26DOI: 10.1515/bmt-2021-0307
Micha Bischofberger, Stephan Böhringer, Erik Schkommodau
This paper proposes a conceptual method to calculate the pose of a stereo-vision camera relative to an artificial mandible without additional markers. The general method for marker-free navigation has four steps: 1) parallel image acquisition by a stereo-vision camera, 2) automatic identification of 2d point pairs (landmark pairs) in a left and a right image, 3) calculation of related 3d points in the joint camera coordinate system and 4) matching of 3d points generated to a preoperative 3d model (i.e., CT data based). To identify and compare landmarks in the acquired stereo images, well-known algorithms for landmark detection, description and matching were compared within the developed approach. Finally, the BRISK algorithm (Leutenegger S, Chli M, Siegwart RY. BRISK: Binary Robust invariant scalable keypoints. Proceedings of the IEEE International Conference on Computer Vision; 2011: 2548-2555) was used. The proposed method was implemented in MATLAB® and validated in vitro with one artificial mandible. The accuracy evaluation of the camera positions calculated resulted in an average deviation error of 1.45 mm ± 0.76 mm to the real camera displacement. This value was calculated using only stereo images with over 100 reconstructed landmark pairs each. This provides the basis for marker-free navigation.
本文提出了一种不需要附加标记来计算立体视觉相机相对于人工下颌骨的姿态的概念方法。无标记导航的一般方法有四个步骤:1)由立体视觉相机获取平行图像,2)自动识别左右图像中的二维点对(地标对),3)在联合相机坐标系中计算相关的三维点,4)将生成的三维点与术前三维模型(即基于CT数据)进行匹配。为了识别和比较获得的立体图像中的地标,在开发的方法中比较了已知的地标检测,描述和匹配算法。Leutenegger S, Chli M, Siegwart RY. BRISK:二值鲁棒不变可伸缩关键点。IEEE计算机视觉国际会议论文集;2011: 2548-2555)。在MATLAB®中实现了该方法,并在一个人工下颌骨上进行了体外验证。计算得到的摄像机位置精度评价结果与摄像机实际位移的平均偏差为1.45 mm±0.76 mm。该值仅使用具有100多个重建地标对的立体图像来计算。这为无标记导航提供了基础。
{"title":"Automatic landmark identification for surgical 3d-navigation - A proposed method for marker-free dental surgical navigation systems.","authors":"Micha Bischofberger, Stephan Böhringer, Erik Schkommodau","doi":"10.1515/bmt-2021-0307","DOIUrl":"https://doi.org/10.1515/bmt-2021-0307","url":null,"abstract":"<p><p>This paper proposes a conceptual method to calculate the pose of a stereo-vision camera relative to an artificial mandible without additional markers. The general method for marker-free navigation has four steps: 1) parallel image acquisition by a stereo-vision camera, 2) automatic identification of 2d point pairs (landmark pairs) in a left and a right image, 3) calculation of related 3d points in the joint camera coordinate system and 4) matching of 3d points generated to a preoperative 3d model (i.e., CT data based). To identify and compare landmarks in the acquired stereo images, well-known algorithms for landmark detection, description and matching were compared within the developed approach. Finally, the BRISK algorithm (Leutenegger S, Chli M, Siegwart RY. BRISK: Binary Robust invariant scalable keypoints. Proceedings of the IEEE International Conference on Computer Vision; 2011: 2548-2555) was used. The proposed method was implemented in MATLAB<sup>®</sup> and validated <i>in vitro</i> with one artificial mandible. The accuracy evaluation of the camera positions calculated resulted in an average deviation error of 1.45 mm ± 0.76 mm to the real camera displacement. This value was calculated using only stereo images with over 100 reconstructed landmark pairs each. This provides the basis for marker-free navigation.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40479492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-28Print Date: 2022-10-26DOI: 10.1515/bmt-2021-0418
Li Ji, Chen Zhang, Haiwei Li, Ningning Zhang, Peng Zheng, Changhao Guo, Yong Zhang, Xiaoyu Tang
The take-off and landing phases are considered the critical stages of aircraft flight. To ensure flight efficiency and safety in the critical stages, this research proposes a method for analyzing and monitoring pilot flight status by β-wave. The focus of the study is β potential changes on the EEG map. First, the proportion of β-wave in the electroencephalogram (EEG) of pilots during take-off and landing increases significantly. Second, the EEG map accurately and intuitively reflects the spatial distribution of potential changes in brain regions. Finally, correlation and machine learning are used for further research of β-wave. The conclusions show that the significant changes in the β-wave caused by take-off and landing can be seen in the EEG map to identify and adjust the pilot's state. Therefore, this research provides more accurate and effective reference information (like the EEG map, correlation and machine learning) for efficient and safe flight training in the critical stages.
{"title":"Analysis of pilots' EEG map in take-off and landing tasks.","authors":"Li Ji, Chen Zhang, Haiwei Li, Ningning Zhang, Peng Zheng, Changhao Guo, Yong Zhang, Xiaoyu Tang","doi":"10.1515/bmt-2021-0418","DOIUrl":"https://doi.org/10.1515/bmt-2021-0418","url":null,"abstract":"<p><p>The take-off and landing phases are considered the critical stages of aircraft flight. To ensure flight efficiency and safety in the critical stages, this research proposes a method for analyzing and monitoring pilot flight status by β-wave. The focus of the study is β potential changes on the EEG map. First, the proportion of β-wave in the electroencephalogram (EEG) of pilots during take-off and landing increases significantly. Second, the EEG map accurately and intuitively reflects the spatial distribution of potential changes in brain regions. Finally, correlation and machine learning are used for further research of β-wave. The conclusions show that the significant changes in the β-wave caused by take-off and landing can be seen in the EEG map to identify and adjust the pilot's state. Therefore, this research provides more accurate and effective reference information (like the EEG map, correlation and machine learning) for efficient and safe flight training in the critical stages.</p>","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40408695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Supervised automatic sleep scoring algorithms are usually trained using sleep stage labels manually annotated on 30 s epochs of PSG data. In this study, we investigate the impact of using shorter epochs with various PSG input signals for training and testing a Long Short Term Memory (LSTM) neural network. An LSTM model is evaluated on the provided 30 s epoch sleep stage labels from a publicly available dataset, as well as on 10 s subdivisions. Additionally, three independent scorers re-labeled a subset of the dataset on shorter time windows. The automatic sleep scoring experiments were repeated on the re-annotated subset.The highest performance is achieved on features extracted from 30 s epochs of a single channel frontal EEG. The resulting accuracy, precision and recall were of 92.22%, 67.58% and 66.00% respectively. When using a shorter epoch as input, the performance decreased by approximately 20%. Re-annotating a subset of the dataset on shorter time epochs did not improve the results and further altered the sleep stage detection performance. Our results show that our feature-based LSTM classification algorithm performs better on 30 s PSG epochs when compared to 10 s epochs used as input. Future work could be oriented to determining whether varying the epoch size improves classification outcomes for different types of classification algorithms.
有监督的自动睡眠评分算法通常使用人工标注睡眠阶段标签的PSG数据进行训练。在这项研究中,我们研究了使用不同PSG输入信号的较短时间对长短期记忆(LSTM)神经网络的训练和测试的影响。LSTM模型在公共数据集提供的30秒epoch睡眠阶段标签以及10秒细分上进行评估。此外,三个独立的评分者在较短的时间窗口上重新标记数据集的子集。在重新标注的子集上重复自动睡眠评分实验。在单通道额叶脑电图的30 s epoch特征提取上,获得了最高的性能。准确度、精密度和召回率分别为92.22%、67.58%和66.00%。当使用较短的epoch作为输入时,性能下降了大约20%。在较短的时间点上重新标注数据集的子集并没有改善结果,反而进一步改变了睡眠阶段检测性能。结果表明,基于特征的LSTM分类算法在30秒的PSG epoch上的表现优于10秒的PSG epoch。未来的工作可能是确定不同epoch大小是否会改善不同类型分类算法的分类结果。
{"title":"Automatic sleep scoring with LSTM networks: impact of time granularity and input signals","authors":"Alexandra-Maria Tăuțan, A. C. Rossi, B. Ionescu","doi":"10.1515/bmt-2021-0408","DOIUrl":"https://doi.org/10.1515/bmt-2021-0408","url":null,"abstract":"Abstract Supervised automatic sleep scoring algorithms are usually trained using sleep stage labels manually annotated on 30 s epochs of PSG data. In this study, we investigate the impact of using shorter epochs with various PSG input signals for training and testing a Long Short Term Memory (LSTM) neural network. An LSTM model is evaluated on the provided 30 s epoch sleep stage labels from a publicly available dataset, as well as on 10 s subdivisions. Additionally, three independent scorers re-labeled a subset of the dataset on shorter time windows. The automatic sleep scoring experiments were repeated on the re-annotated subset.The highest performance is achieved on features extracted from 30 s epochs of a single channel frontal EEG. The resulting accuracy, precision and recall were of 92.22%, 67.58% and 66.00% respectively. When using a shorter epoch as input, the performance decreased by approximately 20%. Re-annotating a subset of the dataset on shorter time epochs did not improve the results and further altered the sleep stage detection performance. Our results show that our feature-based LSTM classification algorithm performs better on 30 s PSG epochs when compared to 10 s epochs used as input. Future work could be oriented to determining whether varying the epoch size improves classification outcomes for different types of classification algorithms.","PeriodicalId":8900,"journal":{"name":"Biomedical Engineering / Biomedizinische Technik","volume":null,"pages":null},"PeriodicalIF":1.7,"publicationDate":"2022-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87029110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}