Pub Date : 2016-06-14DOI: 10.1109/BSN.2016.7516272
O. Binsch, T. Wabeke, P. Valk
Leveraging miniaturized sensor and monitoring technology integrated in easy-to-wear wristband wearables represents a great opportunity for advancing Resilience and Mental Health of e.g. employees that experience high workload. Therefore, it is important to gain insights into the reliability of such technology before far reaching conclusions can be drawn and interventions can be developed. To that aim, we tested three wearable wristband sensor systems (Apple Watch, Microsoft Band and Fitbit Surge) and compared the assessed sensor output with a reliable ground truth. The results showed that heart rate, steps and distance varies considerably around the ground truth during tasks that required body movement. However, during the rest condition (sitting on chair) the heart rate was considered more reliable. It is concluded that caution is warranted while using and interpreting physiological data assessed by the new technology, but, in rest (e.g. pauses, sleep) the wearable' sensors could be used to detect undesirable physiological patterns, indicative of threats to resilience or (mental) health.
{"title":"Comparison of three different physiological wristband sensor systems and their applicability for resilience- and work load monitoring","authors":"O. Binsch, T. Wabeke, P. Valk","doi":"10.1109/BSN.2016.7516272","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516272","url":null,"abstract":"Leveraging miniaturized sensor and monitoring technology integrated in easy-to-wear wristband wearables represents a great opportunity for advancing Resilience and Mental Health of e.g. employees that experience high workload. Therefore, it is important to gain insights into the reliability of such technology before far reaching conclusions can be drawn and interventions can be developed. To that aim, we tested three wearable wristband sensor systems (Apple Watch, Microsoft Band and Fitbit Surge) and compared the assessed sensor output with a reliable ground truth. The results showed that heart rate, steps and distance varies considerably around the ground truth during tasks that required body movement. However, during the rest condition (sitting on chair) the heart rate was considered more reliable. It is concluded that caution is warranted while using and interpreting physiological data assessed by the new technology, but, in rest (e.g. pauses, sleep) the wearable' sensors could be used to detect undesirable physiological patterns, indicative of threats to resilience or (mental) health.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131873374","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516238
B. Groh, Martin Fleckenstein, B. Eskofier
Digital motion analysis in freestyle snowboarding requires a stable trick detection and accurate classification. Freestyle snowboarding contains several trick categories that all have to be recognized for an application in training sessions or competitions. While previous work already addressed the classification of specific tricks or turns, there is no known method that contains a full pipeline for detection and classification of tricks from multiple categories. In this paper, we suggest a classification pipeline containing the detection, categorization and classification of tricks of two major freestyle trick categories. We evaluated our algorithm based on data from two different acquisitions with a total number of eleven athletes and 275 trick events. Tricks of both categories were categorized with recall results of 96.6% and 97.4%. The classification of the tricks was evaluated to an accuracy of 90.3 % for the first and 93.3% for the second category.
{"title":"Wearable trick classification in freestyle snowboarding","authors":"B. Groh, Martin Fleckenstein, B. Eskofier","doi":"10.1109/BSN.2016.7516238","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516238","url":null,"abstract":"Digital motion analysis in freestyle snowboarding requires a stable trick detection and accurate classification. Freestyle snowboarding contains several trick categories that all have to be recognized for an application in training sessions or competitions. While previous work already addressed the classification of specific tricks or turns, there is no known method that contains a full pipeline for detection and classification of tricks from multiple categories. In this paper, we suggest a classification pipeline containing the detection, categorization and classification of tricks of two major freestyle trick categories. We evaluated our algorithm based on data from two different acquisitions with a total number of eleven athletes and 275 trick events. Tricks of both categories were categorized with recall results of 96.6% and 97.4%. The classification of the tricks was evaluated to an accuracy of 90.3 % for the first and 93.3% for the second category.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129354380","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516265
Ebrahim Nemati, Y. Suh, B. Moatamed, M. Sarrafzadeh
A gait velocity estimation algorithm using the inertial sensors of a smartwatch is proposed. The peaks of accelerometer and gyroscope norms are detected at first. Then a Kalman Filter is employed to recover the peaks that are missed because of the arm swing. The Kalman filter combines the accelerometer and gyroscope norm peaks and robustly detect walking step events even in cases where there is a large arm swing. Walking velocity is then estimated using the step duration. It will be shown in this work that the gait velocity has a good correlation with the inverse of the square of the step duration. The model parameters are calculated by collecting the training data from 25 subjects: each subject walked 50 m six times with different walking speed and different arm swing speed. The standard deviation of walking velocity estimation error is 0.1009 m/s (without person dependent calibration) and 0.0630 m/s (with person dependent calibration). The average precision of 91.7% was achieved for the gait speed testing on the smartwatch platform over all the speed scenarios.
{"title":"Gait velocity estimation for a smartwatch platform using Kalman filter peak recovery","authors":"Ebrahim Nemati, Y. Suh, B. Moatamed, M. Sarrafzadeh","doi":"10.1109/BSN.2016.7516265","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516265","url":null,"abstract":"A gait velocity estimation algorithm using the inertial sensors of a smartwatch is proposed. The peaks of accelerometer and gyroscope norms are detected at first. Then a Kalman Filter is employed to recover the peaks that are missed because of the arm swing. The Kalman filter combines the accelerometer and gyroscope norm peaks and robustly detect walking step events even in cases where there is a large arm swing. Walking velocity is then estimated using the step duration. It will be shown in this work that the gait velocity has a good correlation with the inverse of the square of the step duration. The model parameters are calculated by collecting the training data from 25 subjects: each subject walked 50 m six times with different walking speed and different arm swing speed. The standard deviation of walking velocity estimation error is 0.1009 m/s (without person dependent calibration) and 0.0630 m/s (with person dependent calibration). The average precision of 91.7% was achieved for the gait speed testing on the smartwatch platform over all the speed scenarios.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115669325","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516252
B. Moatamed, Arjun, Farhad Shahmohammadi, Ramin Ramezani, A. Naeim, M. Sarrafzadeh
The advent of smart infrastructure or Internet of Things (IoT) has enabled scenarios in which objects with unique identifiers can communicate and transfer data over a network without human to human/computer interactions. Incorporating hardware in such networks is so cheap that it has opened the possibility of connecting just about anything from simple nodes to complex, remotely-monitored sensor networks. In the paper, we describe a low-cost scalable and potentially ubiquitous system for indoor remote health monitoring using low energy bluetooth beacons and a smartwatch. Our system was implemented in a rehabilitation facility in Los Angeles and the overall assessments revealed promising results.
{"title":"Low-cost indoor health monitoring system","authors":"B. Moatamed, Arjun, Farhad Shahmohammadi, Ramin Ramezani, A. Naeim, M. Sarrafzadeh","doi":"10.1109/BSN.2016.7516252","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516252","url":null,"abstract":"The advent of smart infrastructure or Internet of Things (IoT) has enabled scenarios in which objects with unique identifiers can communicate and transfer data over a network without human to human/computer interactions. Incorporating hardware in such networks is so cheap that it has opened the possibility of connecting just about anything from simple nodes to complex, remotely-monitored sensor networks. In the paper, we describe a low-cost scalable and potentially ubiquitous system for indoor remote health monitoring using low energy bluetooth beacons and a smartwatch. Our system was implemented in a rehabilitation facility in Los Angeles and the overall assessments revealed promising results.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125487207","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516223
Joachim F. Kreutzer, J. Deist, C. M. Hein, T. Lüth
Dehydration is a common and severe diagnosis especially among the elderly. Monitoring a healthy fluid intake is therefore vital. In this contribution five sensor approaches that detect fluid intake are presented and compared. Four of the sensor system use an indirect method by monitoring filling levels in a cup. The first concept is equipped with a conductivity based sensor which uses distinct electrodes at different heights for localizing the border between liquid and air. The second system exploits gravity to measure weight changes via hydrostatic pressure at the cup's bottom. In another design the beverage's weight is focused on a force sensor by a mechanism with a movable plate which is separated from the liquid by a flexible foil. The fourth concept uses shielded capacitive sensor detects the capacity inside the cup which is influenced by present media. The final approach monitors the actual fluid intake directly by means of flow measurements compactly integrated into a drinking straw. The implemented system uses a turbine flow meter with two Hall sensors in order to detect passing volume and the direction of the flow. Two electrodes distinguish between air and fluid in order to only monitor beverage intake. Finally, all five sensor designs are evaluated and compared with regard to accuracy, specific restrictions and conceptual realization. Although each concept has distinctive disadvantages they are suitable for detecting filling levels or fluid intake, respectively. A combination of direct and indirect methods to monitor drinking behavior is expected to help prevent dehydrations.
{"title":"Sensor systems for monitoring fluid intake indirectly and directly","authors":"Joachim F. Kreutzer, J. Deist, C. M. Hein, T. Lüth","doi":"10.1109/BSN.2016.7516223","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516223","url":null,"abstract":"Dehydration is a common and severe diagnosis especially among the elderly. Monitoring a healthy fluid intake is therefore vital. In this contribution five sensor approaches that detect fluid intake are presented and compared. Four of the sensor system use an indirect method by monitoring filling levels in a cup. The first concept is equipped with a conductivity based sensor which uses distinct electrodes at different heights for localizing the border between liquid and air. The second system exploits gravity to measure weight changes via hydrostatic pressure at the cup's bottom. In another design the beverage's weight is focused on a force sensor by a mechanism with a movable plate which is separated from the liquid by a flexible foil. The fourth concept uses shielded capacitive sensor detects the capacity inside the cup which is influenced by present media. The final approach monitors the actual fluid intake directly by means of flow measurements compactly integrated into a drinking straw. The implemented system uses a turbine flow meter with two Hall sensors in order to detect passing volume and the direction of the flow. Two electrodes distinguish between air and fluid in order to only monitor beverage intake. Finally, all five sensor designs are evaluated and compared with regard to accuracy, specific restrictions and conceptual realization. Although each concept has distinctive disadvantages they are suitable for detecting filling levels or fluid intake, respectively. A combination of direct and indirect methods to monitor drinking behavior is expected to help prevent dehydrations.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126815085","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516242
Aaron Castillo, Graciela Cortez, David Diaz, Rayton Espiritu, Krystle Ilisastigui, Bryce O'Bard, K. George
The headset mouse is an assistive technology created for individuals with limited to no mobility in their arms. Specifically, this device was created for persons with Amyotrophic Lateral Sclerosis (ALS) also known as Lou Gehrig's disease. The design utilizes a NeuroSky headset, which is used by reading EMG signals to implement mouse clicks using hard blinks and eyebrow raises. A gyroscope is used to read in the values created by the user's head movement and translate that into mouse movement. After creating a prototype device, we were able to test it on both healthy subjects, and persons with ALS (PALS). The PALS had varying neck mobility, with differing progressions of the disease. All subjects were asked to perform four different tasks on a Windows PC that included testing the mouse movement and clicking. Feedback from PALS during testing was used to modify the device in order to better suit their needs. After the four different tasks were conducted with healthy subjects versus PALS, the results showed that most PALS were able to complete the given tasks. Their times of completion were not far off from their healthy counterparts.
{"title":"Hands free mouse","authors":"Aaron Castillo, Graciela Cortez, David Diaz, Rayton Espiritu, Krystle Ilisastigui, Bryce O'Bard, K. George","doi":"10.1109/BSN.2016.7516242","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516242","url":null,"abstract":"The headset mouse is an assistive technology created for individuals with limited to no mobility in their arms. Specifically, this device was created for persons with Amyotrophic Lateral Sclerosis (ALS) also known as Lou Gehrig's disease. The design utilizes a NeuroSky headset, which is used by reading EMG signals to implement mouse clicks using hard blinks and eyebrow raises. A gyroscope is used to read in the values created by the user's head movement and translate that into mouse movement. After creating a prototype device, we were able to test it on both healthy subjects, and persons with ALS (PALS). The PALS had varying neck mobility, with differing progressions of the disease. All subjects were asked to perform four different tasks on a Windows PC that included testing the mouse movement and clicking. Feedback from PALS during testing was used to modify the device in order to better suit their needs. After the four different tasks were conducted with healthy subjects versus PALS, the results showed that most PALS were able to complete the given tasks. Their times of completion were not far off from their healthy counterparts.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115126525","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516291
S. Lee, J. Daneault, Luc Weydert
To circumvent current limitations of wearable sensors that can be used to assess and monitor joint movements, we developed an accurate, low-cost, flexible wearable sensor comprising a retractable reel, a string, and a potentiometer. This sensor is intended to estimate joint angles in correlation with the amount of skin stretch measured by the change in the length of the string. In this study, we validated the accuracy of the sensor against an optoelectronic system in estimating knee joint angles using a dataset obtained from 9 healthy individuals while they walk and run on a treadmill. By our simple calibration procedure, we could convert the voltage output of the potentiometer to the amount of skin stretch as subjects flex or extend their knee. Then, we incorporated a simple polynomial fitting model to estimate the joint angle. Using a leave-one-subject-out cross validation, we achieved an average root mean square error of 4.51 degrees. This work demonstrates the accuracy of the proposed system in estimating knee joint angles and provides the basis to develop more complex systems to assess and monitor joints having more degrees of freedom. We believe that our novel low-cost wearable sensing technology has great potential to enable joint kinematic monitoring in ambulatory settings.
{"title":"A novel flexible wearable sensor for estimating joint-angles","authors":"S. Lee, J. Daneault, Luc Weydert","doi":"10.1109/BSN.2016.7516291","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516291","url":null,"abstract":"To circumvent current limitations of wearable sensors that can be used to assess and monitor joint movements, we developed an accurate, low-cost, flexible wearable sensor comprising a retractable reel, a string, and a potentiometer. This sensor is intended to estimate joint angles in correlation with the amount of skin stretch measured by the change in the length of the string. In this study, we validated the accuracy of the sensor against an optoelectronic system in estimating knee joint angles using a dataset obtained from 9 healthy individuals while they walk and run on a treadmill. By our simple calibration procedure, we could convert the voltage output of the potentiometer to the amount of skin stretch as subjects flex or extend their knee. Then, we incorporated a simple polynomial fitting model to estimate the joint angle. Using a leave-one-subject-out cross validation, we achieved an average root mean square error of 4.51 degrees. This work demonstrates the accuracy of the proposed system in estimating knee joint angles and provides the basis to develop more complex systems to assess and monitor joints having more degrees of freedom. We believe that our novel low-cost wearable sensing technology has great potential to enable joint kinematic monitoring in ambulatory settings.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134091616","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516246
Nick Merrill, Max T. Curran, Jong-Kai Yang, J. Chuang
While brain-computer interfaces (BCI) based on electroencephalography (EEG) have improved dramatically over the past five years, their inconvenient, head-worn form factor has challenged their wider adoption. In this paper, we investigate how EEG signals collected from the ear could be used for “gestural” control of a brain-computer interface (BCI). Specifically, we investigate the efficacy of a support vector classifier (SVC) in distinguishing between mental tasks, or gestures, recorded by a modified, consumer headset. We find that an SVC reaches acceptable BCI accuracy for nine of the subjects in our pool (n=12), and distinguishes at least one pair of gestures better than chance for all subjects. User surveys highlight the need for longer-term research on user attitudes toward in-ear EEG devices, for discreet, non-invasive BCIs.
{"title":"Classifying mental gestures with in-ear EEG","authors":"Nick Merrill, Max T. Curran, Jong-Kai Yang, J. Chuang","doi":"10.1109/BSN.2016.7516246","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516246","url":null,"abstract":"While brain-computer interfaces (BCI) based on electroencephalography (EEG) have improved dramatically over the past five years, their inconvenient, head-worn form factor has challenged their wider adoption. In this paper, we investigate how EEG signals collected from the ear could be used for “gestural” control of a brain-computer interface (BCI). Specifically, we investigate the efficacy of a support vector classifier (SVC) in distinguishing between mental tasks, or gestures, recorded by a modified, consumer headset. We find that an SVC reaches acceptable BCI accuracy for nine of the subjects in our pool (n=12), and distinguishes at least one pair of gestures better than chance for all subjects. User surveys highlight the need for longer-term research on user attitudes toward in-ear EEG devices, for discreet, non-invasive BCIs.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133109628","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516288
Arindam Dutta, O. Ma, M. Buman, D. Bliss
Recent popular emphasis on exercise for personal wellbeing has created a demand for techniques which monitor and classify human activities. Previous studies have shown promising results in applying various classification and feature extraction methods for identifying unique physical activities on various datasets. We apply learning techniques to GENEactiv accelerometer recordings to identify and monitor a wide range of daily activities. The dataset is composed of 92 participants, of ages 20-65, performing 25 unique activities, both ambulatory and non-ambulatory. The algorithm identified 130 different time and frequency domain features and selected the most efficient features with the sequential forward selection algorithm. With classification in two stages with both Gaussian mixture model (GMM) and hidden Markov model (HMM) we have combined the activities with similar features. We have also shown a comparative study between the two classifiers. We achieved an accuracy of 95.5% while classifying 10 unique activities with HMM and 89.7% while classifying 9. The most efficient result is obtained using HMM in 2-D feature space, where it is able to classify 15 unique activities at an accuracy of 90.12%.
{"title":"Learning approach for classification of GENEActiv accelerometer data for unique activity identification","authors":"Arindam Dutta, O. Ma, M. Buman, D. Bliss","doi":"10.1109/BSN.2016.7516288","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516288","url":null,"abstract":"Recent popular emphasis on exercise for personal wellbeing has created a demand for techniques which monitor and classify human activities. Previous studies have shown promising results in applying various classification and feature extraction methods for identifying unique physical activities on various datasets. We apply learning techniques to GENEactiv accelerometer recordings to identify and monitor a wide range of daily activities. The dataset is composed of 92 participants, of ages 20-65, performing 25 unique activities, both ambulatory and non-ambulatory. The algorithm identified 130 different time and frequency domain features and selected the most efficient features with the sequential forward selection algorithm. With classification in two stages with both Gaussian mixture model (GMM) and hidden Markov model (HMM) we have combined the activities with similar features. We have also shown a comparative study between the two classifiers. We achieved an accuracy of 95.5% while classifying 10 unique activities with HMM and 89.7% while classifying 9. The most efficient result is obtained using HMM in 2-D feature space, where it is able to classify 15 unique activities at an accuracy of 90.12%.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133771219","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 : 2016-06-14DOI: 10.1109/BSN.2016.7516262
P. Paladugu, Alejandra Hernandez, Karlie Gross, Yi-Cherng Su, Ahmet Neseli, Sara P. Gombatto, K. Moon, Yusuf Öztürk
Several different factors have been proposed to contribute to the development of chronic low back pain (LBP). Specifically, researchers and clinicians have proposed that impairments of low back posture and movement, particularly during functional activities, are important to address during intervention. However, objective measures of posture and movement are typically only measured in the laboratory setting. Observation of posture and movement in laboratory is limited because people with LBP may not perform naturally when they are being observed, and observation in a single session does not provide information about the duration of postures or frequency of movements across the day. In this paper, we present a wireless body sensor cluster formed by up to seven sensors in order to monitor spine posture and movement both in absolute and relative coordinate systems. The Body Kinematics Monitoring (BKM) system measures the magnitude and frequency of spine movements, and duration of spine postures in 3D, without impeding natural movement. The BKM node developed in this study is 3.0cm in diameter, and contains a 9-axis motion processor that records the raw inertial information of different spine regions. The system offers a standard Bluetooth Low Energy (BLE) protocol to communicate with mobile or fixed hosts. The BKM system has been validated in the laboratory by measuring lumbar spine postures on a mechanical spine testing platform across a known range of angles.
{"title":"A sensor cluster to monitor body kinematics","authors":"P. Paladugu, Alejandra Hernandez, Karlie Gross, Yi-Cherng Su, Ahmet Neseli, Sara P. Gombatto, K. Moon, Yusuf Öztürk","doi":"10.1109/BSN.2016.7516262","DOIUrl":"https://doi.org/10.1109/BSN.2016.7516262","url":null,"abstract":"Several different factors have been proposed to contribute to the development of chronic low back pain (LBP). Specifically, researchers and clinicians have proposed that impairments of low back posture and movement, particularly during functional activities, are important to address during intervention. However, objective measures of posture and movement are typically only measured in the laboratory setting. Observation of posture and movement in laboratory is limited because people with LBP may not perform naturally when they are being observed, and observation in a single session does not provide information about the duration of postures or frequency of movements across the day. In this paper, we present a wireless body sensor cluster formed by up to seven sensors in order to monitor spine posture and movement both in absolute and relative coordinate systems. The Body Kinematics Monitoring (BKM) system measures the magnitude and frequency of spine movements, and duration of spine postures in 3D, without impeding natural movement. The BKM node developed in this study is 3.0cm in diameter, and contains a 9-axis motion processor that records the raw inertial information of different spine regions. The system offers a standard Bluetooth Low Energy (BLE) protocol to communicate with mobile or fixed hosts. The BKM system has been validated in the laboratory by measuring lumbar spine postures on a mechanical spine testing platform across a known range of angles.","PeriodicalId":205735,"journal":{"name":"2016 IEEE 13th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133899555","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：