Pub Date : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014792
V. Abbaraju, K. Lewis, S. Majerus
Analyzing urodynamic study (UDS) tracings can be prone to error in the presence of artifacts and subjective due to lack of standardization in clinical UDS interpretation. As such, the diagnosis of patients undergoing UDS would greatly benefit from a standardized, automated method to assist clinicians in interpreting UDS tracings. In this work, we evaluated a machine learning framework for automatically classifying bladder events from single-channel vesical pressure recordings $(P_{VES}) (N=60)$ into 4 possible classes: abdominal event (i.e., cough or Valsalva), voiding contraction, detrusor overactivity (DO) and no event. Wavelet multiresolution analysis of $P_{VES}$ was used to extract time-frequency localized wavelet coefficient vectors which were segmented into 0.8 second segments with 55 statistical features per segment. Feature selection was subsequently applied for three classifier architectures: a k-nearest classifier (KNN), an artificial neural network classifier (ANN) and a support vector machine classifier (SVM). Each classifier was trained and evaluated using five-fold cross validation, from which we derived the sensitivity, specificity, F1 score and AUC for all four classes and the overall classification accuracy for each classifier. The KNN, ANN and SVM classifiers labeled 7,861 0.8 second $P_{VES}$ segments with 91.5%, 90.8% and 82.4% accuracy, respectively. We have thus proposed the first framework for automated multi-event bladder classification using single-channel UDS data.
{"title":"Machine Learning for Automated Bladder Event Classification from Single-Channel Vesical Pressure Recordings","authors":"V. Abbaraju, K. Lewis, S. Majerus","doi":"10.1109/SPMB55497.2022.10014792","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014792","url":null,"abstract":"Analyzing urodynamic study (UDS) tracings can be prone to error in the presence of artifacts and subjective due to lack of standardization in clinical UDS interpretation. As such, the diagnosis of patients undergoing UDS would greatly benefit from a standardized, automated method to assist clinicians in interpreting UDS tracings. In this work, we evaluated a machine learning framework for automatically classifying bladder events from single-channel vesical pressure recordings $(P_{VES}) (N=60)$ into 4 possible classes: abdominal event (i.e., cough or Valsalva), voiding contraction, detrusor overactivity (DO) and no event. Wavelet multiresolution analysis of $P_{VES}$ was used to extract time-frequency localized wavelet coefficient vectors which were segmented into 0.8 second segments with 55 statistical features per segment. Feature selection was subsequently applied for three classifier architectures: a k-nearest classifier (KNN), an artificial neural network classifier (ANN) and a support vector machine classifier (SVM). Each classifier was trained and evaluated using five-fold cross validation, from which we derived the sensitivity, specificity, F1 score and AUC for all four classes and the overall classification accuracy for each classifier. The KNN, ANN and SVM classifiers labeled 7,861 0.8 second $P_{VES}$ segments with 91.5%, 90.8% and 82.4% accuracy, respectively. We have thus proposed the first framework for automated multi-event bladder classification using single-channel UDS data.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130471587","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014954
S. Goerttler, M. Wu, F. He
Multivariate signals are signals consisting of multiple signals measured simultaneously over time and are most commonly acquired by sensor networks. The emerging field of graph signal processing (GSP) promises to analyse dynamic characteristics of multivariate signals, while at the same time taking the network, or spatial structure between the signals into account. To do so, GSP decomposes the multivariate signals into graph frequency signals, which are ordered by their magnitude. However, the meaning of the graph frequencies in terms of this ordering remains poorly understood. Here, we investigate the role the ordering plays in preserving valuable dynamic structures in the signals, with neuroimaging applications in mind. In order to overcome the limitations in sample size common to neurophysiological data sets, we introduce a minimalist simulation framework to generate arbitrary amounts of data. Using this artificial data, we find that lower graph frequency signals are less suitable for classifying neurophysiological data than higher graph frequency signals. We further introduce a baseline testing framework for GSP. Using this framework, we conclude that dynamic, or spectral structures are poorly preserved in GSP, high-lighting current limitations of GSP for neuroimaging.
{"title":"The Effect of Graph Frequencies on Dynamic Structures in Graph Signal Processing","authors":"S. Goerttler, M. Wu, F. He","doi":"10.1109/SPMB55497.2022.10014954","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014954","url":null,"abstract":"Multivariate signals are signals consisting of multiple signals measured simultaneously over time and are most commonly acquired by sensor networks. The emerging field of graph signal processing (GSP) promises to analyse dynamic characteristics of multivariate signals, while at the same time taking the network, or spatial structure between the signals into account. To do so, GSP decomposes the multivariate signals into graph frequency signals, which are ordered by their magnitude. However, the meaning of the graph frequencies in terms of this ordering remains poorly understood. Here, we investigate the role the ordering plays in preserving valuable dynamic structures in the signals, with neuroimaging applications in mind. In order to overcome the limitations in sample size common to neurophysiological data sets, we introduce a minimalist simulation framework to generate arbitrary amounts of data. Using this artificial data, we find that lower graph frequency signals are less suitable for classifying neurophysiological data than higher graph frequency signals. We further introduce a baseline testing framework for GSP. Using this framework, we conclude that dynamic, or spectral structures are poorly preserved in GSP, high-lighting current limitations of GSP for neuroimaging.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126414890","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014948
Antoine Honoré, H. Siren, R. Vinuesa, S. Chatterjee, E. Herlenius
Early and accurate neonatal sepsis detection (NSD) can help reduce mortality, morbidity and antibiotic consumption in premature infants. NSD models are often designed and evaluated in case control setups and using data derived from patient electrocardiogram (ECG) only. In this article, we evaluate our models in a more realistic retrospective cohort study setup. We use data from different modalities, including ECG, chest impedance, pulse oximetry, demographics factors and repetitive measurements of body weights. We study both the vanilla and long-short-term-memory (LSTM) Recurrent Neural Networks (RNN) architectures in a sequence to sequence mapping framework for NSD. We compare the performances of the models with logistic regression (LR) on a variety of classification metrics in a leave-one-out cross validation framework. The population we used contains 118 very low birth weight infants, among which 10 experienced sepsis. We showed that LSTM-based RNNs are both (1) more conservative and (2) more precise than LR or vanilla RNN, with a true negative rate at least +26% higher and a precision score of 0.16 compared to 0.06 for LR. This indicates that LSTM-based RNNs have the potential to reduce the false alarm rate of existing linear models, while providing a reliable diagnostic aid for neonatal sepsis.
{"title":"An LSTM-based Recurrent Neural Network for Neonatal Sepsis Detection in Preterm Infants","authors":"Antoine Honoré, H. Siren, R. Vinuesa, S. Chatterjee, E. Herlenius","doi":"10.1109/SPMB55497.2022.10014948","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014948","url":null,"abstract":"Early and accurate neonatal sepsis detection (NSD) can help reduce mortality, morbidity and antibiotic consumption in premature infants. NSD models are often designed and evaluated in case control setups and using data derived from patient electrocardiogram (ECG) only. In this article, we evaluate our models in a more realistic retrospective cohort study setup. We use data from different modalities, including ECG, chest impedance, pulse oximetry, demographics factors and repetitive measurements of body weights. We study both the vanilla and long-short-term-memory (LSTM) Recurrent Neural Networks (RNN) architectures in a sequence to sequence mapping framework for NSD. We compare the performances of the models with logistic regression (LR) on a variety of classification metrics in a leave-one-out cross validation framework. The population we used contains 118 very low birth weight infants, among which 10 experienced sepsis. We showed that LSTM-based RNNs are both (1) more conservative and (2) more precise than LR or vanilla RNN, with a true negative rate at least +26% higher and a precision score of 0.16 compared to 0.06 for LR. This indicates that LSTM-based RNNs have the potential to reduce the false alarm rate of existing linear models, while providing a reliable diagnostic aid for neonatal sepsis.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121362414","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014785
A. Emadi, A. Abdi
It is well understood that the CREB protein is highly involved in neuronal mechanisms underlying memory and learning in mammalian brain, and deficiencies in CREB activity can result in transition to certain pathological conditions. In this paper, we use some published experimental data, along with a neuronal system composed of the Izhikevich neuron model, to characterize how CREB abnormalities can alter neuronal signals and the system behavior. The abnormal data are extracted from intracellular recordings collected from the neurons of transgenic mice expressing VP16-CREB - a constitutively active form of CREB - whereas the normal data are obtained from the wild-type mice neurons. Upon estimating the neuron model parameters from the experimental data, we observe that the model exhibits good fit to both normal and abnormal data, for various synaptic input currents. To study the effect of CREB abnormalities on the considered neuronal system, we use the information theoretic redundancy parameter. It basically measures - for the system output neuron - the amount of spike count information overlap that exists between the states of the stimulus currents injected to the input neurons. Our analysis reveals a noticeable increase in the information redundancy, when CREB behaves abnormally. This finding motivates further exploration of the biological implications of the information redundancy in neuronal systems, and its use as a parameter to model abnormalities in CREB and perhaps other important transcription factors involved in learning and memory.
{"title":"A Study of How Abnormalities of the CREB Protein Affect a Neuronal System and Its Signals: Modeling and Analysis Using Experimental Data","authors":"A. Emadi, A. Abdi","doi":"10.1109/SPMB55497.2022.10014785","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014785","url":null,"abstract":"It is well understood that the CREB protein is highly involved in neuronal mechanisms underlying memory and learning in mammalian brain, and deficiencies in CREB activity can result in transition to certain pathological conditions. In this paper, we use some published experimental data, along with a neuronal system composed of the Izhikevich neuron model, to characterize how CREB abnormalities can alter neuronal signals and the system behavior. The abnormal data are extracted from intracellular recordings collected from the neurons of transgenic mice expressing VP16-CREB - a constitutively active form of CREB - whereas the normal data are obtained from the wild-type mice neurons. Upon estimating the neuron model parameters from the experimental data, we observe that the model exhibits good fit to both normal and abnormal data, for various synaptic input currents. To study the effect of CREB abnormalities on the considered neuronal system, we use the information theoretic redundancy parameter. It basically measures - for the system output neuron - the amount of spike count information overlap that exists between the states of the stimulus currents injected to the input neurons. Our analysis reveals a noticeable increase in the information redundancy, when CREB behaves abnormally. This finding motivates further exploration of the biological implications of the information redundancy in neuronal systems, and its use as a parameter to model abnormalities in CREB and perhaps other important transcription factors involved in learning and memory.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134402891","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014966
D. D. Kairamkonda, P. S. Mandaleeka, A. Favaro, C. Motley, A. Butala, E. Oh, R. Stevens, N. Dehak, L. Moro-Velázquez
Clinicians currently use handwriting as one of the methods to establish the presence and monitor the progression of neurodegenerative diseases (NDs). While common handwriting evaluation methods are valuable means to detect fine motor and cognitive impairments associated with NDs, these are observer-dependent and subjective. In the present study, we analyzed a broad array of interpretable features, some proposed for the first time in this study, obtained from online handwriting data of participants with NDs and control subjects (CTRL). ND participants have Alzheimer's disease (AD), Parkinson's disease (PD), or Parkinson's disease mimics (PDM). Hand-writing data from three different neuropsychological tasks was used: Copy Text task, Copy Cube task, and Copy Image task. Then, we arranged three complementary sets of features and conducted a statistical analysis to test their significance between groups. Overall results suggested that subjects with AD reported a significantly higher $(p < 0.05)$ amount of data points and total duration with respect to the CTRL group in almost all the tasks under assessment. On the other hand, subjects with PD showed a significantly lower $(p < 0.05)$ horizontal width (both on tablet and in the air). Even though the AD and PDM groups showed a significantly lower velocity and acceleration $(p < 0.05)$, their number of inversions in velocity and acceleration was significantly greater $(p < 0.05)$, which indicates disfluency in writing. The features that we have used were found to provide good results in differentiating the studied groups and could be considered as part of diagnostic tools for the assessment and monitoring of NDs in clinical trials.
{"title":"Analysis of Interpretable Handwriting Features to Evaluate Motoric Patterns in Different Neurodegenerative Diseases","authors":"D. D. Kairamkonda, P. S. Mandaleeka, A. Favaro, C. Motley, A. Butala, E. Oh, R. Stevens, N. Dehak, L. Moro-Velázquez","doi":"10.1109/SPMB55497.2022.10014966","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014966","url":null,"abstract":"Clinicians currently use handwriting as one of the methods to establish the presence and monitor the progression of neurodegenerative diseases (NDs). While common handwriting evaluation methods are valuable means to detect fine motor and cognitive impairments associated with NDs, these are observer-dependent and subjective. In the present study, we analyzed a broad array of interpretable features, some proposed for the first time in this study, obtained from online handwriting data of participants with NDs and control subjects (CTRL). ND participants have Alzheimer's disease (AD), Parkinson's disease (PD), or Parkinson's disease mimics (PDM). Hand-writing data from three different neuropsychological tasks was used: Copy Text task, Copy Cube task, and Copy Image task. Then, we arranged three complementary sets of features and conducted a statistical analysis to test their significance between groups. Overall results suggested that subjects with AD reported a significantly higher $(p < 0.05)$ amount of data points and total duration with respect to the CTRL group in almost all the tasks under assessment. On the other hand, subjects with PD showed a significantly lower $(p < 0.05)$ horizontal width (both on tablet and in the air). Even though the AD and PDM groups showed a significantly lower velocity and acceleration $(p < 0.05)$, their number of inversions in velocity and acceleration was significantly greater $(p < 0.05)$, which indicates disfluency in writing. The features that we have used were found to provide good results in differentiating the studied groups and could be considered as part of diagnostic tools for the assessment and monitoring of NDs in clinical trials.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115957684","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014799
B. Buttaro
Bacterial biofilms are a ubiquitous form of bacterial growth. Biofilms consisting of bacterial, viruses, and protozoa exist in the environment and our gastrointestinal tract. Robust bacterial biofilms surviving off light and fixing carbon exist in deserts and on marble monuments. Biofilms can be a medical challenge when composed of a single pathogenic species or promote human health as a highly evolved microbiome ecology composed of hundreds of species. Regardless of their location, certain patterns emerge. Biofilms can behave as viscous liquids or rigid structures. The rigid structures can provide protection to more viscous biofilms. These structures can be intricately organized structures ready to respond to changes in environmental conditions at a moment's notice. Within complex multi-species communities, bacteria organize themselves into smaller communities, which are often interdependent of each other. Understanding the rules that govern their structural and spatial arrangements supporting their functions and interactions is a complex challenge best approached by a multidisciplinary approach of computational mathematics, mathematical modeling, machine learning, and engineering. In this talk, we will review biofilm basics defining what is a bacterial biofilm, their ubiquitous nature, and their roles in promoting health and producing difficult to treat diseases. Then we will explore processes shared by biofilms independent of their environment and specific bacterial species composition and the methods to study them. Studying these processes may reveal underlying principles driving the structural and spatial arrangements of most biofilms. Topics will include the composition and material properties of biofilms and how ordered matrix molecules, and possibly aggregation, contribute to rigid structuredevelopment. The next part of the talk will review the function of rigid structures. Rigid structures form when bacteria are under stress, including antibiotic stress, to provide protection to the community allowing survival and even continued growth. This suggests multicellular behavior with parts of the community providing protection to other regions that are actively growing to replace the dying cells resulting in steady state survival of a community including the formation of regions of viscous biofilm behind rigid structures under flow. Thiswill include a discussion on how mobile genetic elements can reshape biofilms and possibly make commensal microbiota bacteria more pathogenic (able to cause disease). Finally, the use of simple interdependent communities in extreme environments will be discussed as a model for spatial organization of biofilms communities, which may have implications for establishment of interdependent smaller communities within the context of larger multi-kingdom species.
{"title":"Playing by the Rules: Structural and Spatial Organization of Biofilm Communities","authors":"B. Buttaro","doi":"10.1109/SPMB55497.2022.10014799","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014799","url":null,"abstract":"Bacterial biofilms are a ubiquitous form of bacterial growth. Biofilms consisting of bacterial, viruses, and protozoa exist in the environment and our gastrointestinal tract. Robust bacterial biofilms surviving off light and fixing carbon exist in deserts and on marble monuments. Biofilms can be a medical challenge when composed of a single pathogenic species or promote human health as a highly evolved microbiome ecology composed of hundreds of species. Regardless of their location, certain patterns emerge. Biofilms can behave as viscous liquids or rigid structures. The rigid structures can provide protection to more viscous biofilms. These structures can be intricately organized structures ready to respond to changes in environmental conditions at a moment's notice. Within complex multi-species communities, bacteria organize themselves into smaller communities, which are often interdependent of each other. Understanding the rules that govern their structural and spatial arrangements supporting their functions and interactions is a complex challenge best approached by a multidisciplinary approach of computational mathematics, mathematical modeling, machine learning, and engineering. In this talk, we will review biofilm basics defining what is a bacterial biofilm, their ubiquitous nature, and their roles in promoting health and producing difficult to treat diseases. Then we will explore processes shared by biofilms independent of their environment and specific bacterial species composition and the methods to study them. Studying these processes may reveal underlying principles driving the structural and spatial arrangements of most biofilms. Topics will include the composition and material properties of biofilms and how ordered matrix molecules, and possibly aggregation, contribute to rigid structuredevelopment. The next part of the talk will review the function of rigid structures. Rigid structures form when bacteria are under stress, including antibiotic stress, to provide protection to the community allowing survival and even continued growth. This suggests multicellular behavior with parts of the community providing protection to other regions that are actively growing to replace the dying cells resulting in steady state survival of a community including the formation of regions of viscous biofilm behind rigid structures under flow. Thiswill include a discussion on how mobile genetic elements can reshape biofilms and possibly make commensal microbiota bacteria more pathogenic (able to cause disease). Finally, the use of simple interdependent communities in extreme environments will be discussed as a model for spatial organization of biofilms communities, which may have implications for establishment of interdependent smaller communities within the context of larger multi-kingdom species.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129007217","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014961
N. Sadowski, G. Drzewiecki
The sinoatrial node (SAN), located in the right atrium wall, is the heart's biological pacemaker and determines heart rate due to the repetitive spontaneous action potentials for cardiac rhythmic contractions in the heart pacemaker cells. The funny current (If) and SAN, together with regulation by the sympathetic and parasympathetic nervous systems, modulate the frequency of the SAN action potential. The interaction of these systems is responsible for the rhythmic pacemaker activity, controlling heart rate, and abnormalities resulting in arrhythmias.
{"title":"Computational Cellular Model of Heart Rate Variability During Controlled Respiration","authors":"N. Sadowski, G. Drzewiecki","doi":"10.1109/SPMB55497.2022.10014961","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014961","url":null,"abstract":"The sinoatrial node (SAN), located in the right atrium wall, is the heart's biological pacemaker and determines heart rate due to the repetitive spontaneous action potentials for cardiac rhythmic contractions in the heart pacemaker cells. The funny current (If) and SAN, together with regulation by the sympathetic and parasympathetic nervous systems, modulate the frequency of the SAN action potential. The interaction of these systems is responsible for the rhythmic pacemaker activity, controlling heart rate, and abnormalities resulting in arrhythmias.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129250497","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014868
A. Borovac, T. Runarsson, G. Thorvardsson, S. Gudmundsson
An EEG seizure detection algorithm employed in a clinical setting is likely to encounter many EEG segments that are difficult to classify due to the complexity of EEG signals and small data sets frequently used to train seizure detectors. The detectors should therefore be able to notify the clinician when they are uncertain in their predictions and they should also be accurate for confident predictions. This would enable the clinician to focus mainly on the parts of the recording where confidence in predictions is low. Here we analyse the calibration of neonatal and adult seizure detection algorithms based on a convolutional neural network in terms of how well the output seizure/non-seizure probabilities estimate the corresponding empirical frequencies. We found that the detectors turned out to be overconfident, in particular when incorrectly predicting seizure segments as non-seizure segments. The calibration of both detectors, measured in terms of expected calibration error and overconfidence error, was improved noticeably with the use of Monte Carlo dropout. We find that a straightforward application of dropout during training and classification leads to a noticeable improvement in the calibration of EEG seizure detectors based on a convolutional neural network.
{"title":"Calibration of Automatic Seizure Detection Algorithms","authors":"A. Borovac, T. Runarsson, G. Thorvardsson, S. Gudmundsson","doi":"10.1109/SPMB55497.2022.10014868","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014868","url":null,"abstract":"An EEG seizure detection algorithm employed in a clinical setting is likely to encounter many EEG segments that are difficult to classify due to the complexity of EEG signals and small data sets frequently used to train seizure detectors. The detectors should therefore be able to notify the clinician when they are uncertain in their predictions and they should also be accurate for confident predictions. This would enable the clinician to focus mainly on the parts of the recording where confidence in predictions is low. Here we analyse the calibration of neonatal and adult seizure detection algorithms based on a convolutional neural network in terms of how well the output seizure/non-seizure probabilities estimate the corresponding empirical frequencies. We found that the detectors turned out to be overconfident, in particular when incorrectly predicting seizure segments as non-seizure segments. The calibration of both detectors, measured in terms of expected calibration error and overconfidence error, was improved noticeably with the use of Monte Carlo dropout. We find that a straightforward application of dropout during training and classification leads to a noticeable improvement in the calibration of EEG seizure detectors based on a convolutional neural network.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"13 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114103405","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014741
A. Subramanian, F. Shamsi, L. Najafizadeh
An important application of brain computer interface devices (BCIs) is in assistive systems for patients with motor and communication disabilities. Due to their condition, these patients may experience pain. However, how the presence of pain influences the operation of such BCIs has not been fully investigated. This paper studies the impact of the presence of acute pain on the classification accuracy of a BCI, which employs functional near infrared spectroscopy (fNIRS) for brain signal acquisition. Cortical signals are obtained in the presence and absence of an external pain stimulus, while participants perform two mental arithmetic tasks. Convolutional neural network (CNN) is used to classify the tasks. It is observed that when the classifier is trained on pain-free data and tested on data obtained in the presence of pain, the classification accuracy significantly drops. Next, multi-label classification is performed to simultaneously identify the presence of pain and classify the tasks, further demonstrating that the distinction of tasks in the presence of pain is challenging. Finally, to mitigate the impact of pain, it is proposed to train the model collectively on data obtained in the presence and the absence of pain. It is observed that using this approach significantly improves the classification accuracy. Our results suggest that it is critical to include data obtained in the presence of pain in the training process of the classification models, when designing BCIs in assistive systems for patients.
{"title":"Analysis of the Impact of the Presence of Physical Pain in fNIRS-based BCI Systems","authors":"A. Subramanian, F. Shamsi, L. Najafizadeh","doi":"10.1109/SPMB55497.2022.10014741","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014741","url":null,"abstract":"An important application of brain computer interface devices (BCIs) is in assistive systems for patients with motor and communication disabilities. Due to their condition, these patients may experience pain. However, how the presence of pain influences the operation of such BCIs has not been fully investigated. This paper studies the impact of the presence of acute pain on the classification accuracy of a BCI, which employs functional near infrared spectroscopy (fNIRS) for brain signal acquisition. Cortical signals are obtained in the presence and absence of an external pain stimulus, while participants perform two mental arithmetic tasks. Convolutional neural network (CNN) is used to classify the tasks. It is observed that when the classifier is trained on pain-free data and tested on data obtained in the presence of pain, the classification accuracy significantly drops. Next, multi-label classification is performed to simultaneously identify the presence of pain and classify the tasks, further demonstrating that the distinction of tasks in the presence of pain is challenging. Finally, to mitigate the impact of pain, it is proposed to train the model collectively on data obtained in the presence and the absence of pain. It is observed that using this approach significantly improves the classification accuracy. Our results suggest that it is critical to include data obtained in the presence of pain in the training process of the classification models, when designing BCIs in assistive systems for patients.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124395322","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 : 2022-12-03DOI: 10.1109/SPMB55497.2022.10014869
M. Huang, E. Clancy
Hip fractures are common in the geriatric population and represent a growing social and economic burden [1]. They are associated with decreased mobility, and the recovery period can be prolonged. Early mobilization is a critical component of the recovery process. Few studies have quantitatively measured activity levels in patients after hip fracture surgery, resulting in a lack of objective data about mobility status after hospital discharge. Furthermore, each year about 1.5 million elderly people are injured falling, and about 47,300 people aged ≥65 years suffer injuries from falls using walking aids that require an emergency room visit [2]. Patients using walkers are seven times more likely to fall than those that use canes. The risk of repeat falls has been shown to be especially high in patients who have already sustained a hip fracture.
{"title":"Smart Walker: an IMU-Based Device for Patient Activity Logging and Fall Detection","authors":"M. Huang, E. Clancy","doi":"10.1109/SPMB55497.2022.10014869","DOIUrl":"https://doi.org/10.1109/SPMB55497.2022.10014869","url":null,"abstract":"Hip fractures are common in the geriatric population and represent a growing social and economic burden [1]. They are associated with decreased mobility, and the recovery period can be prolonged. Early mobilization is a critical component of the recovery process. Few studies have quantitatively measured activity levels in patients after hip fracture surgery, resulting in a lack of objective data about mobility status after hospital discharge. Furthermore, each year about 1.5 million elderly people are injured falling, and about 47,300 people aged ≥65 years suffer injuries from falls using walking aids that require an emergency room visit [2]. Patients using walkers are seven times more likely to fall than those that use canes. The risk of repeat falls has been shown to be especially high in patients who have already sustained a hip fracture.","PeriodicalId":261445,"journal":{"name":"2022 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129528908","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：