Conductive elastomer elements can be industrially embedded into garments to form unobtrusive stra... more Conductive elastomer elements can be industrially embedded into garments to form unobtrusive strain sensing stripes. The present article outlines the structure of a strain-sensor based gesture detection algorithm. Current sensing prototypes include several dozens of sensors; their redundancy with respect to the limb's degrees of freedom, and other artifacts implied by this measurement technique, call for the development of novel robust multivariate pattern-matching techniques. The algorithm's construction is explained, and its performances are evaluated in the context of motor rehabilitation exercises for both two-class and multi-class tasks.
The purpose of this study was to assess the performance of a real-time (&... more The purpose of this study was to assess the performance of a real-time ("open-end") version of the dynamic time warping (DTW) algorithm for the recognition of motor exercises. Given a possibly incomplete input stream of data and a reference time series, the open-end DTW algorithm computes both the size of the prefix of reference which is best matched by the input, and the dissimilarity between the matched portions. The algorithm was used to provide real-time feedback to neurological patients undergoing motor rehabilitation. We acquired a dataset of multivariate time series from a sensorized long-sleeve shirt which contains 29 strain sensors distributed on the upper limb. Seven typical rehabilitation exercises were recorded in several variations, both correctly and incorrectly executed, and at various speeds, totaling a data set of 840 time series. Nearest-neighbour classifiers were built according to the outputs of open-end DTW alignments and their global counterparts on exercise pairs. The classifiers were also tested on well-known public datasets from heterogeneous domains. Nonparametric tests show that (1) on full time series the two algorithms achieve the same classification accuracy (p-value =0.32); (2) on partial time series, classifiers based on open-end DTW have a far higher accuracy (kappa=0.898 versus kappa=0.447;p<10(-5)); and (3) the prediction of the matched fraction follows closely the ground truth (root mean square <10%). The results hold for the motor rehabilitation and the other datasets tested, as well. The open-end variant of the DTW algorithm is suitable for the classification of truncated quantitative time series, even in the presence of noise. Early recognition and accurate class prediction can be achieved, provided that enough variance is available over the time span of the reference. Therefore, the proposed technique expands the use of DTW to a wider range of applications, such as real-time biofeedback systems.
A multivariate dataset obtained recording the stretch of various segments of the fabric of a sens... more A multivariate dataset obtained recording the stretch of various segments of the fabric of a sensorized garment while executing several rehabilitation exercises. Full description in P. Tormene, T. Giorgino, S. Quaglini, M. Stefanelli. <em>Matching incomplete time series with dynamic time warping: an algorithm and an application to post-stroke rehabilitation.</em> Artificial Intelligence in Medicine, Volume 45, Issue 1, Pages 11-34, http://dx.doi.org/10.1016/j.artmed.2008.11.007 and in enclosed Readme file.
MobiGuide is a project devoted to the development of a patient-centric decision support system ba... more MobiGuide is a project devoted to the development of a patient-centric decision support system based on computerized clinical guidelines for chronic illnesses including Atrial . In this paper we describe the process of
Studies in Health Technology and Informatics, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases' management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart's Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
Studies in Health Technology and Informatics, Feb 1, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39; management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
European journal of physical and rehabilitation medicine
The aim of this study was to provide a description of a newly-developed remote rehabilitation sys... more The aim of this study was to provide a description of a newly-developed remote rehabilitation system that can be employed both at home and in the hospital, supporting motor rehabilitation for post-stroke patients with upper-limb impairment. A garment, which embeds kinesthetic sensors made of a piezoresitive polymer, is provided with a wireless connection to a computer (the patient station). The station detects in real time whether the patient is performing the exercises correctly or not, and provides feedback through an easy visual representation on the screen. Movement recognition is performed using a template matching approach, which allows exercises to be defined during each session as required without additional configuration. In this study an healthy volunteer used the garment to record 840 exercises, mimicking both correct and incorrect compensatory movements under expert supervision. The sensitivity and specificity of the recognition system were measured through its ability to correctly identify the pre-labelled exercises. A pilot set of 13 post-stroke subjects (mean age of 50) was then offered to use the rehabilitation system while in the neuro-rehabilitation ward; the acceptability was assessed through a 10-question subjective evaluation questionnaire. The wearable system tested provided a raw recognition performance (correct-versus-incorrect exercise detection) above 90%. The majority of the patients were satisfied with the system, considered it useful, and would use it at home. In conclusion, computer-based interventions can support widespread, earlier and more intense physical therapy after a neurological event, provided they are easy to use and blend well with the existing rehabilitation workflow. Wearable sensors are promising candidates to realize unobtrusive devices to support the rehabilitation process and its continuity after discharge from the Rehabilitation Unit.
Studies in health technology and informatics, 2013
MobiGuide is a project devoted to the development of a patient-centric decision support system ba... more MobiGuide is a project devoted to the development of a patient-centric decision support system based on computerized clinical guidelines for chronic illnesses including Atrial Fibrillation (AF). In this paper we describe the process of (1) identifying guideline recommendations that will require patients to take actions (e.g., take measurement, take drug), thus impacting patients' daily-life behavior, (2) eliciting from the medical experts the corresponding set of personalized operationalized advices that are not explicitly written in the guideline (patient-tailored workflow patterns) and (3) delivering this advice to patients. The analysis of the AF guideline has resulted in four types of patient-tailored workflow patterns: therapy-related advisors, measurements advisors, suggestions for dealing with interventions that may require modulating patient therapy, and personalized packages for close monitoring of patients. We will show how these patterns can be generated using informa...
Studies in health technology and informatics, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases' management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart's Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
ABSTRACT The purpose of this paper is to outline the methodology and workflow of the Integrated R... more ABSTRACT The purpose of this paper is to outline the methodology and workflow of the Integrated Risk Modelling Toolkit and database of the Global Earthquake Model (GEM). The Integrated Risk Modelling Toolkit was developed to allow users to meaningfully integrate quantitative assessments of social and economic conditions that affect impacts and loss with physical risk estimates for earthquakes. The Integrated Risk Modelling Toolkit for integrated risk assessment is being designed to allow user-data inputs enabling high resolution and bottom-up quantitative analyses (e.g. local-level analysis that is participatory and regionally and context specific) for the development of social, economic, and integrated risk indices. Moreover, it is intended that the software tool will allow users manipulate and interact with all of the data, which may be inputs to both physical earthquake risk and social vulnerability models.
IEEE Transactions on Information Technology in Biomedicine, 2000
Recent studies suggest that the quality of recovery after a stroke can be increased by early and ... more Recent studies suggest that the quality of recovery after a stroke can be increased by early and more intensive rehabilitation. Portable unobtrusive devices are promising candidates for the realization of stroke-rehabilitation systems that complement care in the post-acute rehabilitation phase, both in the clinic and at home. The proposed system allows patients to increase the amount of motor exercise they can perform in autonomy, providing them with a real-time feedback based on wearable sensors embedded in the garment&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s tissue across the upper limb and trunk. A dynamic time warping algorithm allows for the recognition of correct and incorrect motor exercises. After the feedback phase, data are stored in a central location for review and statistics. Workstations can be installed either at home or at the hospital to support patients, independent of their location. The performance of the system on healthy subjects was quantified for seven types of motor exercises for upper limb&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s rehabilitation. Finally, we present the preliminary results of a pilot clinical study to test the system&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s acceptability and usability by post-stroke patients, and physicians in a clinical setting.
The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;... more The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;open-end&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;) version of the dynamic time warping (DTW) algorithm for the recognition of motor exercises. Given a possibly incomplete input stream of data and a reference time series, the open-end DTW algorithm computes both the size of the prefix of reference which is best matched by the input, and the dissimilarity between the matched portions. The algorithm was used to provide real-time feedback to neurological patients undergoing motor rehabilitation. We acquired a dataset of multivariate time series from a sensorized long-sleeve shirt which contains 29 strain sensors distributed on the upper limb. Seven typical rehabilitation exercises were recorded in several variations, both correctly and incorrectly executed, and at various speeds, totaling a data set of 840 time series. Nearest-neighbour classifiers were built according to the outputs of open-end DTW alignments and their global counterparts on exercise pairs. The classifiers were also tested on well-known public datasets from heterogeneous domains. Nonparametric tests show that (1) on full time series the two algorithms achieve the same classification accuracy (p-value =0.32); (2) on partial time series, classifiers based on open-end DTW have a far higher accuracy (kappa=0.898 versus kappa=0.447;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10(-5)); and (3) the prediction of the matched fraction follows closely the ground truth (root mean square &amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10%). The results hold for the motor rehabilitation and the other datasets tested, as well. The open-end variant of the DTW algorithm is suitable for the classification of truncated quantitative time series, even in the presence of noise. Early recognition and accurate class prediction can be achieved, provided that enough variance is available over the time span of the reference. Therefore, the proposed technique expands the use of DTW to a wider range of applications, such as real-time biofeedback systems.
Conductive elastomer elements can be industrially embedded into garments to form unobtrusive stra... more Conductive elastomer elements can be industrially embedded into garments to form unobtrusive strain sensing stripes. The present article outlines the structure of a strain-sensor based gesture detection algorithm. Current sensing prototypes include several dozens of sensors; their redundancy with respect to the limb&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s degrees of freedom, and other artifacts implied by this measurement technique, call for the development of novel robust multivariate pattern-matching techniques. The algorithm&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s construction is explained, and its performances are evaluated in the context of motor rehabilitation exercises for both two-class and multi-class tasks.
The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;... more The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;open-end&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;) version of the dynamic time warping (DTW) algorithm for the recognition of motor exercises. Given a possibly incomplete input stream of data and a reference time series, the open-end DTW algorithm computes both the size of the prefix of reference which is best matched by the input, and the dissimilarity between the matched portions. The algorithm was used to provide real-time feedback to neurological patients undergoing motor rehabilitation. We acquired a dataset of multivariate time series from a sensorized long-sleeve shirt which contains 29 strain sensors distributed on the upper limb. Seven typical rehabilitation exercises were recorded in several variations, both correctly and incorrectly executed, and at various speeds, totaling a data set of 840 time series. Nearest-neighbour classifiers were built according to the outputs of open-end DTW alignments and their global counterparts on exercise pairs. The classifiers were also tested on well-known public datasets from heterogeneous domains. Nonparametric tests show that (1) on full time series the two algorithms achieve the same classification accuracy (p-value =0.32); (2) on partial time series, classifiers based on open-end DTW have a far higher accuracy (kappa=0.898 versus kappa=0.447;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10(-5)); and (3) the prediction of the matched fraction follows closely the ground truth (root mean square &amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10%). The results hold for the motor rehabilitation and the other datasets tested, as well. The open-end variant of the DTW algorithm is suitable for the classification of truncated quantitative time series, even in the presence of noise. Early recognition and accurate class prediction can be achieved, provided that enough variance is available over the time span of the reference. Therefore, the proposed technique expands the use of DTW to a wider range of applications, such as real-time biofeedback systems.
A multivariate dataset obtained recording the stretch of various segments of the fabric of a sens... more A multivariate dataset obtained recording the stretch of various segments of the fabric of a sensorized garment while executing several rehabilitation exercises. Full description in P. Tormene, T. Giorgino, S. Quaglini, M. Stefanelli. <em>Matching incomplete time series with dynamic time warping: an algorithm and an application to post-stroke rehabilitation.</em> Artificial Intelligence in Medicine, Volume 45, Issue 1, Pages 11-34, http://dx.doi.org/10.1016/j.artmed.2008.11.007 and in enclosed Readme file.
MobiGuide is a project devoted to the development of a patient-centric decision support system ba... more MobiGuide is a project devoted to the development of a patient-centric decision support system based on computerized clinical guidelines for chronic illnesses including Atrial . In this paper we describe the process of
Studies in Health Technology and Informatics, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases' management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart's Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
Studies in Health Technology and Informatics, Feb 1, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39; management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
European journal of physical and rehabilitation medicine
The aim of this study was to provide a description of a newly-developed remote rehabilitation sys... more The aim of this study was to provide a description of a newly-developed remote rehabilitation system that can be employed both at home and in the hospital, supporting motor rehabilitation for post-stroke patients with upper-limb impairment. A garment, which embeds kinesthetic sensors made of a piezoresitive polymer, is provided with a wireless connection to a computer (the patient station). The station detects in real time whether the patient is performing the exercises correctly or not, and provides feedback through an easy visual representation on the screen. Movement recognition is performed using a template matching approach, which allows exercises to be defined during each session as required without additional configuration. In this study an healthy volunteer used the garment to record 840 exercises, mimicking both correct and incorrect compensatory movements under expert supervision. The sensitivity and specificity of the recognition system were measured through its ability to correctly identify the pre-labelled exercises. A pilot set of 13 post-stroke subjects (mean age of 50) was then offered to use the rehabilitation system while in the neuro-rehabilitation ward; the acceptability was assessed through a 10-question subjective evaluation questionnaire. The wearable system tested provided a raw recognition performance (correct-versus-incorrect exercise detection) above 90%. The majority of the patients were satisfied with the system, considered it useful, and would use it at home. In conclusion, computer-based interventions can support widespread, earlier and more intense physical therapy after a neurological event, provided they are easy to use and blend well with the existing rehabilitation workflow. Wearable sensors are promising candidates to realize unobtrusive devices to support the rehabilitation process and its continuity after discharge from the Rehabilitation Unit.
Studies in health technology and informatics, 2013
MobiGuide is a project devoted to the development of a patient-centric decision support system ba... more MobiGuide is a project devoted to the development of a patient-centric decision support system based on computerized clinical guidelines for chronic illnesses including Atrial Fibrillation (AF). In this paper we describe the process of (1) identifying guideline recommendations that will require patients to take actions (e.g., take measurement, take drug), thus impacting patients' daily-life behavior, (2) eliciting from the medical experts the corresponding set of personalized operationalized advices that are not explicitly written in the guideline (patient-tailored workflow patterns) and (3) delivering this advice to patients. The analysis of the AF guideline has resulted in four types of patient-tailored workflow patterns: therapy-related advisors, measurements advisors, suggestions for dealing with interventions that may require modulating patient therapy, and personalized packages for close monitoring of patients. We will show how these patterns can be generated using informa...
Studies in health technology and informatics, 2007
Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare co... more Stroke is a serious neurological accident which accounts for a wide fraction of the healthcare costs in industrialized societies. Recovery from stroke and other neurological accidents usually include motor rehabilitation, maintained for several months, and logopedic training for the recovery of cognitive and speech abilities. The MyHeart consortium is addressing several aspects of cardiovascular diseases' management by combining clothes with embedded biomedical sensors and information technologies. One of the application areas is especially devoted to supporting Neurological Rehabilitation (NR). This article describes how MyHeart's Concept NR is structured and how technologies are leveraged to support both motor rehabilitation and speech/cognitive training. Information technology and garment-embedded sensors, combined, permit assisted training both at the clinic and at home, after discharge from the intensive care unit.
ABSTRACT The purpose of this paper is to outline the methodology and workflow of the Integrated R... more ABSTRACT The purpose of this paper is to outline the methodology and workflow of the Integrated Risk Modelling Toolkit and database of the Global Earthquake Model (GEM). The Integrated Risk Modelling Toolkit was developed to allow users to meaningfully integrate quantitative assessments of social and economic conditions that affect impacts and loss with physical risk estimates for earthquakes. The Integrated Risk Modelling Toolkit for integrated risk assessment is being designed to allow user-data inputs enabling high resolution and bottom-up quantitative analyses (e.g. local-level analysis that is participatory and regionally and context specific) for the development of social, economic, and integrated risk indices. Moreover, it is intended that the software tool will allow users manipulate and interact with all of the data, which may be inputs to both physical earthquake risk and social vulnerability models.
IEEE Transactions on Information Technology in Biomedicine, 2000
Recent studies suggest that the quality of recovery after a stroke can be increased by early and ... more Recent studies suggest that the quality of recovery after a stroke can be increased by early and more intensive rehabilitation. Portable unobtrusive devices are promising candidates for the realization of stroke-rehabilitation systems that complement care in the post-acute rehabilitation phase, both in the clinic and at home. The proposed system allows patients to increase the amount of motor exercise they can perform in autonomy, providing them with a real-time feedback based on wearable sensors embedded in the garment&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s tissue across the upper limb and trunk. A dynamic time warping algorithm allows for the recognition of correct and incorrect motor exercises. After the feedback phase, data are stored in a central location for review and statistics. Workstations can be installed either at home or at the hospital to support patients, independent of their location. The performance of the system on healthy subjects was quantified for seven types of motor exercises for upper limb&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s rehabilitation. Finally, we present the preliminary results of a pilot clinical study to test the system&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;#39;s acceptability and usability by post-stroke patients, and physicians in a clinical setting.
The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;... more The purpose of this study was to assess the performance of a real-time (&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;open-end&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;) version of the dynamic time warping (DTW) algorithm for the recognition of motor exercises. Given a possibly incomplete input stream of data and a reference time series, the open-end DTW algorithm computes both the size of the prefix of reference which is best matched by the input, and the dissimilarity between the matched portions. The algorithm was used to provide real-time feedback to neurological patients undergoing motor rehabilitation. We acquired a dataset of multivariate time series from a sensorized long-sleeve shirt which contains 29 strain sensors distributed on the upper limb. Seven typical rehabilitation exercises were recorded in several variations, both correctly and incorrectly executed, and at various speeds, totaling a data set of 840 time series. Nearest-neighbour classifiers were built according to the outputs of open-end DTW alignments and their global counterparts on exercise pairs. The classifiers were also tested on well-known public datasets from heterogeneous domains. Nonparametric tests show that (1) on full time series the two algorithms achieve the same classification accuracy (p-value =0.32); (2) on partial time series, classifiers based on open-end DTW have a far higher accuracy (kappa=0.898 versus kappa=0.447;p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10(-5)); and (3) the prediction of the matched fraction follows closely the ground truth (root mean square &amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;10%). The results hold for the motor rehabilitation and the other datasets tested, as well. The open-end variant of the DTW algorithm is suitable for the classification of truncated quantitative time series, even in the presence of noise. Early recognition and accurate class prediction can be achieved, provided that enough variance is available over the time span of the reference. Therefore, the proposed technique expands the use of DTW to a wider range of applications, such as real-time biofeedback systems.
Uploads
Papers by Paolo Tormene