Pub Date : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962876
R. An, M. N. Hasan, Yuncheng Man, U. Gurkan
Anemia affects more than 2 billion people worldwide, which is about 25% of the world’s population. Anemia has numerous causes ranging from nutritional deficiencies, drugs, chronic conditions that indirectly cause anemia as well as primary hematologic diseases. Among the various causes of anemia world-wide, hemoglobinopathies, including Sickle Cell Disease (SCD) and Thalassemia, are the 3rd most prevalent after iron-deficiency anemia and hookworm disease. Anemia and SCD diagnosis/monitoring are challenging in low and middle income countries due to lack of laboratory infrastructure and skilled personnel as well as insufficient financial resources. We extended our previously established HemeChip system to add total hemoglobin quantification and anemia testing capability. HemeChip+ is mass-producible at low cost and offers the first and only single test point-of-care (POC) platform for portable, affordable, and accurate, hemoglobin quantification, anemia detection, and hemoglobin variant identification.
{"title":"Integrated Point-of-Care Device for Anemia Detection and Hemoglobin Variant Identification","authors":"R. An, M. N. Hasan, Yuncheng Man, U. Gurkan","doi":"10.1109/HI-POCT45284.2019.8962876","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962876","url":null,"abstract":"Anemia affects more than 2 billion people worldwide, which is about 25% of the world’s population. Anemia has numerous causes ranging from nutritional deficiencies, drugs, chronic conditions that indirectly cause anemia as well as primary hematologic diseases. Among the various causes of anemia world-wide, hemoglobinopathies, including Sickle Cell Disease (SCD) and Thalassemia, are the 3rd most prevalent after iron-deficiency anemia and hookworm disease. Anemia and SCD diagnosis/monitoring are challenging in low and middle income countries due to lack of laboratory infrastructure and skilled personnel as well as insufficient financial resources. We extended our previously established HemeChip system to add total hemoglobin quantification and anemia testing capability. HemeChip+ is mass-producible at low cost and offers the first and only single test point-of-care (POC) platform for portable, affordable, and accurate, hemoglobin quantification, anemia detection, and hemoglobin variant identification.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116841208","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962697
M. Sami, Kurt Wagner, P. Parikh, U. Hassan
The architecture and working of a smartphone-based biosensor for the quantification of leukocytes at point of care is presented in this paper. The biosensor consists of a microscopic smartphone attachment with a resolution of 6.2 μm and a disposable microfluidic biochip for capturing leukocytes. Polymorphonuclear leukocytes (PMNL) were isolated from whole blood before being seeded into PBS solution to mimic the biological samples from patients suffering from various diseases. To capture all the leukocytes, antihuman CD45 antibody was immobilized in the capture chamber of microfluidic biochip for one hour for adsorption. Leukocyte spiked PBS sample was then flowed through the microfluidic biochip at 10 μl/min for capturing leukocytes. 50 μl of a green nuclear stain was then flowed through the biochip for fluorescent imaging. Leukocyte capture was verified by imaging the biochip in the smartphone setup. ImageJ was then used for detection and quantification of leukocytes from the captured images. The obtained results showcase the feasibility of this setup for detection of multiple biomarkers from different body fluids at point of care.
{"title":"Smartphone Based Microfluidic Biosensor for Leukocyte Quantification at the Point-of-Care","authors":"M. Sami, Kurt Wagner, P. Parikh, U. Hassan","doi":"10.1109/HI-POCT45284.2019.8962697","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962697","url":null,"abstract":"The architecture and working of a smartphone-based biosensor for the quantification of leukocytes at point of care is presented in this paper. The biosensor consists of a microscopic smartphone attachment with a resolution of 6.2 μm and a disposable microfluidic biochip for capturing leukocytes. Polymorphonuclear leukocytes (PMNL) were isolated from whole blood before being seeded into PBS solution to mimic the biological samples from patients suffering from various diseases. To capture all the leukocytes, antihuman CD45 antibody was immobilized in the capture chamber of microfluidic biochip for one hour for adsorption. Leukocyte spiked PBS sample was then flowed through the microfluidic biochip at 10 μl/min for capturing leukocytes. 50 μl of a green nuclear stain was then flowed through the biochip for fluorescent imaging. Leukocyte capture was verified by imaging the biochip in the smartphone setup. ImageJ was then used for detection and quantification of leukocytes from the captured images. The obtained results showcase the feasibility of this setup for detection of multiple biomarkers from different body fluids at point of care.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126548907","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 : 2019-11-01DOI: 10.1109/hi-poct45284.2019.8962893
{"title":"HI-POCT 2019 Keynote Speakers","authors":"","doi":"10.1109/hi-poct45284.2019.8962893","DOIUrl":"https://doi.org/10.1109/hi-poct45284.2019.8962893","url":null,"abstract":"","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123599363","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962763
Qingxue Zhang
Smart health is paving a promising way for modern health management. Daily activity and fall risk monitoring is one important application that urges smart technologies, resulting from the fact that there are 29 million falls and 7 million fall injuries per year, and also the fact that appropriate exercise can lower the risk of death by up to 20 to 70%. However, it is very challenging to accurately identify an activity due to the diversity of the human biomechanical dynamics. Main reasons include: even a same person usually has different motion characteristics when performing a same activity; there are many different activities in our daily lives; and the sensor wearing habit may be different. In this paper, focusing on these challenges, a new intelligent computational approach is proposed for robust activity detection, leveraging biomechanical dynamics enhancement and deep learning technologies. It can unveil deep hidden biomechanical patterns from the mobile phone-sensed motion data, and robustly detect 17 types of daily and fall activities performed by 30 people. The detection accuracy of 11,770 activities is as high as 93.9%, indicating the effectiveness of the proposed approach. This research is expected to greatly advance mobile daily activity and fall risk monitoring in smart health era.
{"title":"Deep Learning of Biomechanical Dynamics in Mobile Daily Activity and Fall Risk Monitoring*","authors":"Qingxue Zhang","doi":"10.1109/HI-POCT45284.2019.8962763","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962763","url":null,"abstract":"Smart health is paving a promising way for modern health management. Daily activity and fall risk monitoring is one important application that urges smart technologies, resulting from the fact that there are 29 million falls and 7 million fall injuries per year, and also the fact that appropriate exercise can lower the risk of death by up to 20 to 70%. However, it is very challenging to accurately identify an activity due to the diversity of the human biomechanical dynamics. Main reasons include: even a same person usually has different motion characteristics when performing a same activity; there are many different activities in our daily lives; and the sensor wearing habit may be different. In this paper, focusing on these challenges, a new intelligent computational approach is proposed for robust activity detection, leveraging biomechanical dynamics enhancement and deep learning technologies. It can unveil deep hidden biomechanical patterns from the mobile phone-sensed motion data, and robustly detect 17 types of daily and fall activities performed by 30 people. The detection accuracy of 11,770 activities is as high as 93.9%, indicating the effectiveness of the proposed approach. This research is expected to greatly advance mobile daily activity and fall risk monitoring in smart health era.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123800904","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962742
Qingxue Zhang, Kyle Frick
As a leading cause of death, cardiac diseases are taking away lives from over a half million US people each year. Standard 12-lead electrocardiogram (ECG) signals are gold-standard cardiac vital signs, and have been widely used in clinics and hospitals. However, it is still not readily available in our daily lives, due to its inconvenient and uncomfortable setting, as well as large signal quality degradation during our daily motions. In this research, a novel ECG monitor called, All-ECG, is proposed, which is expected to, at the same time, provide a convenient setting and enable motion-tolerant 12-lead ECG tracking. To achieve the first goal – convenience, a least-number of leads are selected to reconstruct the remaining leads. To achieve the second goal – robustness, a deep learning framework based on the long short-term memory is developed to reconstruct high quality ECG leads from noisy ECG leads. Evaluated on patient ECG data, the proposed deep learning framework can effectively reconstruct standard 12-lead ECG only from noisy 3-lead ECG during daily motions, with a correlation coefficient of as high as 0.82 and a root mean square error of 0.073 mV. To the best of our knowledge, this is the first study on 12-lead ECG reconstruction from a least-number of noisy leads, and is expected to greatly advance long-term daily heart health management.
{"title":"All-ECG: A Least-number of Leads ECG Monitor for Standard 12-lead ECG Tracking during Motion*","authors":"Qingxue Zhang, Kyle Frick","doi":"10.1109/HI-POCT45284.2019.8962742","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962742","url":null,"abstract":"As a leading cause of death, cardiac diseases are taking away lives from over a half million US people each year. Standard 12-lead electrocardiogram (ECG) signals are gold-standard cardiac vital signs, and have been widely used in clinics and hospitals. However, it is still not readily available in our daily lives, due to its inconvenient and uncomfortable setting, as well as large signal quality degradation during our daily motions. In this research, a novel ECG monitor called, All-ECG, is proposed, which is expected to, at the same time, provide a convenient setting and enable motion-tolerant 12-lead ECG tracking. To achieve the first goal – convenience, a least-number of leads are selected to reconstruct the remaining leads. To achieve the second goal – robustness, a deep learning framework based on the long short-term memory is developed to reconstruct high quality ECG leads from noisy ECG leads. Evaluated on patient ECG data, the proposed deep learning framework can effectively reconstruct standard 12-lead ECG only from noisy 3-lead ECG during daily motions, with a correlation coefficient of as high as 0.82 and a root mean square error of 0.073 mV. To the best of our knowledge, this is the first study on 12-lead ECG reconstruction from a least-number of noisy leads, and is expected to greatly advance long-term daily heart health management.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129542714","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962654
Rui Hua, Ya Wang
Motor dysfunction, a well-known early sign of neurodegenerative diseases, is occurring to seniors at a growing rate and affects their physical capability of independent living if not treated effectively. The symptoms of motor dysfunction are hard to notice at early stages and can deteriorate in the long term. Thus, it is desirable to detect motor function changes in daily life in a noninvasive manner. This paper aims to accomplish this goal by proposing a method to auto-recognize nine types of daily activities from continuous movements with the use of a smart insole and a pre-designed route map. The route map creates a semi-controlled environment to help the subjects take actions comfortably and behave in experiments as they do in real life. The nine types of highly similar activities are selected from the motor examination and the balance evaluation system test. Preliminary experiments were done with four subjects with controlled and uncontrolled data collection. Four supervised machine learning classifiers are evaluated and compared for classification performance with a 2s window and different overlaps. With regards to the performance and robustness of classifiers, the Random Forest classifier trained with Mix Dataset shows the best results with an averaged classification accuracy of 98.19% in model training, 92.67% in cross-validation and 83.87% in prediction. The results show that it is feasible to recognize these nine activities from daily locomotor movement and further extract parameters of interest from activity periods for early motor dysfunction detection.
{"title":"Daily Locomotor Movement Recognition with a Smart Insole and a Pre-defined Route Map: Towards Early Motor Dysfunction Detection*","authors":"Rui Hua, Ya Wang","doi":"10.1109/HI-POCT45284.2019.8962654","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962654","url":null,"abstract":"Motor dysfunction, a well-known early sign of neurodegenerative diseases, is occurring to seniors at a growing rate and affects their physical capability of independent living if not treated effectively. The symptoms of motor dysfunction are hard to notice at early stages and can deteriorate in the long term. Thus, it is desirable to detect motor function changes in daily life in a noninvasive manner. This paper aims to accomplish this goal by proposing a method to auto-recognize nine types of daily activities from continuous movements with the use of a smart insole and a pre-designed route map. The route map creates a semi-controlled environment to help the subjects take actions comfortably and behave in experiments as they do in real life. The nine types of highly similar activities are selected from the motor examination and the balance evaluation system test. Preliminary experiments were done with four subjects with controlled and uncontrolled data collection. Four supervised machine learning classifiers are evaluated and compared for classification performance with a 2s window and different overlaps. With regards to the performance and robustness of classifiers, the Random Forest classifier trained with Mix Dataset shows the best results with an averaged classification accuracy of 98.19% in model training, 92.67% in cross-validation and 83.87% in prediction. The results show that it is feasible to recognize these nine activities from daily locomotor movement and further extract parameters of interest from activity periods for early motor dysfunction detection.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116635239","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962811
R. Deo, S. Panigrahi
Diabetes is a major chronic disease which impacts all age groups. It has increasing prevalence worldwide. Certain factors increase the chances of diabetes occurrence in individuals. Prediction-based modeling has been used previously to provide a prevention based approach to diabetes. Prediction models have predominantly been based on regression and feature elimination. In this paper, a machine learning-based approach is presented to predict the individual diabetes occurrence based on specific lifestyle, and demographic factors. A publicly available dataset - continuous NHANES, was used. To account for small data size due to missing data and class imbalanced data, certain statistical techniques were applied. Synthetic minority over sampling technique was used via Gower’s distance calculation to avoid class imbalanced data. Additionally, principal component analysis was used as a feature extraction technique. Predictive models were developed using MATLAB. A dataset with 140 data samples and 11 predictor variables (converted to eight principal components) was used. The output variable had two classes - diabetic and not diabetic. A training data set of 98 and 42 samples for training and testing respectively. Two machine learning models - bagged trees and linear SVM were developed. Two validation techniques - 5- fold cross validation and holdout validation were assessed. The highest accuracy of 91% (90.82%, on test data) was obtained by the linear SVM model using both 5-fold cross validation and hold out validation approaches (AUC of 0.908 in both cases).
{"title":"Performance Assessment of Machine Learning Based Models for Diabetes Prediction","authors":"R. Deo, S. Panigrahi","doi":"10.1109/HI-POCT45284.2019.8962811","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962811","url":null,"abstract":"Diabetes is a major chronic disease which impacts all age groups. It has increasing prevalence worldwide. Certain factors increase the chances of diabetes occurrence in individuals. Prediction-based modeling has been used previously to provide a prevention based approach to diabetes. Prediction models have predominantly been based on regression and feature elimination. In this paper, a machine learning-based approach is presented to predict the individual diabetes occurrence based on specific lifestyle, and demographic factors. A publicly available dataset - continuous NHANES, was used. To account for small data size due to missing data and class imbalanced data, certain statistical techniques were applied. Synthetic minority over sampling technique was used via Gower’s distance calculation to avoid class imbalanced data. Additionally, principal component analysis was used as a feature extraction technique. Predictive models were developed using MATLAB. A dataset with 140 data samples and 11 predictor variables (converted to eight principal components) was used. The output variable had two classes - diabetic and not diabetic. A training data set of 98 and 42 samples for training and testing respectively. Two machine learning models - bagged trees and linear SVM were developed. Two validation techniques - 5- fold cross validation and holdout validation were assessed. The highest accuracy of 91% (90.82%, on test data) was obtained by the linear SVM model using both 5-fold cross validation and hold out validation approaches (AUC of 0.908 in both cases).","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122122945","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962833
F. Tabei, B. Askarian, J. Chong
Health parameters such as heart rhythm, blood pressure, and the level of oxygen saturation in the blood could be measured with photoplethysmography (PPG) signal. The advent of smartphone camera sensors has enabled the extraction of PPG signals from smartphones. PPG signals are weak at motion and noise artifacts (MNA) which could generate unreliable heart rate measurement. Smartphone PPG signals are more prone to MNA since they are not designed for clinical applications. PPG signals are known as biometric signals since they have unique behaviors for each individual. However, in previous MNA detection studies this personalized characteristic has not been considered. In this paper, we propose a novel personalized MNA detection method by applying a probabilistic neural network as a classifier. The performance of our personalized method is evaluated with 25 volunteered subjects in terms of accuracy, specificity, and sensitivity and compared with the generalized method. The average accuracy of our personalized method is 97.96% while it is 92.94% in the generalized one. The average values of personalized specificity and sensitivity are 99.69% and 93.91% while the generalized classifier gives 95.38% and 87.4%.
{"title":"Motion and Noise Artifact Detection in Smartphone Photoplethysmograph Signals Using Personalized Classifier","authors":"F. Tabei, B. Askarian, J. Chong","doi":"10.1109/HI-POCT45284.2019.8962833","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962833","url":null,"abstract":"Health parameters such as heart rhythm, blood pressure, and the level of oxygen saturation in the blood could be measured with photoplethysmography (PPG) signal. The advent of smartphone camera sensors has enabled the extraction of PPG signals from smartphones. PPG signals are weak at motion and noise artifacts (MNA) which could generate unreliable heart rate measurement. Smartphone PPG signals are more prone to MNA since they are not designed for clinical applications. PPG signals are known as biometric signals since they have unique behaviors for each individual. However, in previous MNA detection studies this personalized characteristic has not been considered. In this paper, we propose a novel personalized MNA detection method by applying a probabilistic neural network as a classifier. The performance of our personalized method is evaluated with 25 volunteered subjects in terms of accuracy, specificity, and sensitivity and compared with the generalized method. The average accuracy of our personalized method is 97.96% while it is 92.94% in the generalized one. The average values of personalized specificity and sensitivity are 99.69% and 93.91% while the generalized classifier gives 95.38% and 87.4%.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129765532","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962839
Allison Fellger, Gina Sprint, Alexa Andrews, D. Weeks, Elena Crooks
Actigraphs are wearable sensors used to collect activity and sleep time series data from healthy and unhealthy populations. Unhealthy populations, such as individuals undergoing inpatient rehabilitation, typically exhibit abnormal daytime physical activity and nighttime sleeping patterns due to their injury and drastic changes in their activities of daily living. Consequently, Actigraph data collected from patients attending inpatient rehabilitation are often noisy and can be difficult to reliably draw conclusions from. In this paper, we apply machine learning to analyze such highly variable Actigraph data. We collected 24-hour, minute-by-minute Actigraph data from 17 patients receiving inpatient therapy post-stroke or post-traumatic brain injury. Our approach utilizes similarities among historical sequences of data to train machine learning algorithms to predict nighttime sleep duration. By tuning parameters related to our regression algorithm, we obtained a normalized root mean square error of 14.40%. Our approach is suitable for point of care and remote monitoring to detect changes in sleep for individuals recovering from stroke and traumatic brain injuries.
{"title":"Nighttime Sleep Duration Prediction for Inpatient Rehabilitation Using Similar Actigraphy Sequences","authors":"Allison Fellger, Gina Sprint, Alexa Andrews, D. Weeks, Elena Crooks","doi":"10.1109/HI-POCT45284.2019.8962839","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962839","url":null,"abstract":"Actigraphs are wearable sensors used to collect activity and sleep time series data from healthy and unhealthy populations. Unhealthy populations, such as individuals undergoing inpatient rehabilitation, typically exhibit abnormal daytime physical activity and nighttime sleeping patterns due to their injury and drastic changes in their activities of daily living. Consequently, Actigraph data collected from patients attending inpatient rehabilitation are often noisy and can be difficult to reliably draw conclusions from. In this paper, we apply machine learning to analyze such highly variable Actigraph data. We collected 24-hour, minute-by-minute Actigraph data from 17 patients receiving inpatient therapy post-stroke or post-traumatic brain injury. Our approach utilizes similarities among historical sequences of data to train machine learning algorithms to predict nighttime sleep duration. By tuning parameters related to our regression algorithm, we obtained a normalized root mean square error of 14.40%. Our approach is suitable for point of care and remote monitoring to detect changes in sleep for individuals recovering from stroke and traumatic brain injuries.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126472439","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 : 2019-11-01DOI: 10.1109/HI-POCT45284.2019.8962890
Imran M. Saied, Syed Ali Akbar Hussainy
In recent years, there have been considerable developments in smart wearable devices and unobtrusive monitoring systems that can be used in detecting and monitoring a patient’s health. However, these technological advances have not been implemented for head diagnostics, where the majority of hospitals still relying on MRI or CT scans which are bulky and expensive. In this paper, a wearable and portable device is presented that can be used for microwave head diagnostic systems. The device contains 8 RF sensors that are placed in the inner lining of a hat. The sensors are then connected to a miniaturized vector network analyzer (VNA) that generates and receives signals from the sensors. The signals from the VNA can be captured and processed in a laptop, or it can transfer the data via a Bluetooth module to a mobile device that can process the data in an app. Experiments were performed on a brain phantom to verify the performance of the device. Objects of different sizes were placed in the phantom and measured to represent diseases such as stroke and tumour. Results from the experiments showed that the deice was capable of detecting different levels of diseases in the brain. As a result, the proposed device provides a promising technique for non-invasive head diagnostics that is wearable, portable, and inexpensive.
{"title":"Portable and Wearable Device for Microwave Head Diagnostic Systems","authors":"Imran M. Saied, Syed Ali Akbar Hussainy","doi":"10.1109/HI-POCT45284.2019.8962890","DOIUrl":"https://doi.org/10.1109/HI-POCT45284.2019.8962890","url":null,"abstract":"In recent years, there have been considerable developments in smart wearable devices and unobtrusive monitoring systems that can be used in detecting and monitoring a patient’s health. However, these technological advances have not been implemented for head diagnostics, where the majority of hospitals still relying on MRI or CT scans which are bulky and expensive. In this paper, a wearable and portable device is presented that can be used for microwave head diagnostic systems. The device contains 8 RF sensors that are placed in the inner lining of a hat. The sensors are then connected to a miniaturized vector network analyzer (VNA) that generates and receives signals from the sensors. The signals from the VNA can be captured and processed in a laptop, or it can transfer the data via a Bluetooth module to a mobile device that can process the data in an app. Experiments were performed on a brain phantom to verify the performance of the device. Objects of different sizes were placed in the phantom and measured to represent diseases such as stroke and tumour. Results from the experiments showed that the deice was capable of detecting different levels of diseases in the brain. As a result, the proposed device provides a promising technique for non-invasive head diagnostics that is wearable, portable, and inexpensive.","PeriodicalId":269346,"journal":{"name":"2019 IEEE Healthcare Innovations and Point of Care Technologies, (HI-POCT)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117222180","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：