Development of novel techniques to classify physical activity mode using accelerometers. Pober, D., M., Staudenmayer, J., Raphael, C., & Freedson, P., S. Medicine and science in sports and exercise, 38(9):1626-34, 9, 2006.
Paper
Website abstract bibtex PURPOSE: Use of accelerometers to assess physical activity (PA) is widespread in public health research, but their utility is often limited by the accuracy of data-processing techniques. We hypothesized that more sophisticated approaches to data processing could distinguish between activity types based on accelerometer data, providing a more accurate picture of PA. METHODS: Using data from MTI Actigraphs worn by six subjects during four activities (walking, walking uphill, vacuuming, working at a computer), quadratic discriminant analysis (QDA) was performed, and a hidden Markov model (HMM) was trained to recognize the activities. The ability of the new analytic techniques to accurately classify PA was assessed. RESULTS: The mean (SE) percentage of time points for which the QDA correctly identified activity mode was 70.9% (1.2%). Computer work was correctly recognized most frequently (mean (SE) percent correct = 100% (0.01%)), followed by vacuuming (67.5% (1.5%)), uphill walking (58.2% (3.5%)), and walking (53.6% (3.3%)). The mean (SE) percentage of time points for which the HMM correctly identified activity mode was 80.8% (0.9%). Vacuuming was correctly recognized most frequently (mean (SE) percent correct = 98.8% (0.05%)), followed by computer work (97.3% (0.7%)), walking (62.6% (2.3%)), and uphill walking (62.5% (2.3%)). In contrast to a traditional method of data processing that misidentified the intensity level of 100% of the time spent vacuuming and walking uphill, the QDA and HMM approaches correctly estimated the intensity of activity 99% of the time. CONCLUSION: The novel approach of estimating activity mode, rather than activity level, may allow for more accurate field-based estimates of physical activity using accelerometer data, and this approach warrants more study in a larger and more diverse population of subjects and activities.
@article{
title = {Development of novel techniques to classify physical activity mode using accelerometers.},
type = {article},
year = {2006},
identifiers = {[object Object]},
keywords = {Acceleration,Activities of Daily Living,Adult,Biological,Computer-Assisted,Exercise Test,Exercise Test: classification,Exercise Test: methods,Female,Humans,Male,Markov Chains,Models,Motor Activity,Numerical Analysis,Walking},
pages = {1626-34},
volume = {38},
websites = {http://www.ncbi.nlm.nih.gov/pubmed/16960524},
month = {9},
id = {3dd99bf7-92d0-39a9-a42b-0e80b43b1071},
created = {2016-04-14T18:02:32.000Z},
accessed = {2014-11-04},
file_attached = {true},
profile_id = {d28af011-5164-3af4-8522-822cff4de1eb},
group_id = {e39cd875-9ef8-3fee-ad92-c9d084a63048},
last_modified = {2017-10-14T23:14:07.576Z},
tags = {accelerometry},
read = {true},
starred = {false},
authored = {false},
confirmed = {true},
hidden = {false},
citation_key = {Pober2006},
folder_uuids = {dc34bb65-a794-457d-b88a-9687552166f4},
private_publication = {false},
abstract = {PURPOSE: Use of accelerometers to assess physical activity (PA) is widespread in public health research, but their utility is often limited by the accuracy of data-processing techniques. We hypothesized that more sophisticated approaches to data processing could distinguish between activity types based on accelerometer data, providing a more accurate picture of PA. METHODS: Using data from MTI Actigraphs worn by six subjects during four activities (walking, walking uphill, vacuuming, working at a computer), quadratic discriminant analysis (QDA) was performed, and a hidden Markov model (HMM) was trained to recognize the activities. The ability of the new analytic techniques to accurately classify PA was assessed. RESULTS: The mean (SE) percentage of time points for which the QDA correctly identified activity mode was 70.9% (1.2%). Computer work was correctly recognized most frequently (mean (SE) percent correct = 100% (0.01%)), followed by vacuuming (67.5% (1.5%)), uphill walking (58.2% (3.5%)), and walking (53.6% (3.3%)). The mean (SE) percentage of time points for which the HMM correctly identified activity mode was 80.8% (0.9%). Vacuuming was correctly recognized most frequently (mean (SE) percent correct = 98.8% (0.05%)), followed by computer work (97.3% (0.7%)), walking (62.6% (2.3%)), and uphill walking (62.5% (2.3%)). In contrast to a traditional method of data processing that misidentified the intensity level of 100% of the time spent vacuuming and walking uphill, the QDA and HMM approaches correctly estimated the intensity of activity 99% of the time. CONCLUSION: The novel approach of estimating activity mode, rather than activity level, may allow for more accurate field-based estimates of physical activity using accelerometer data, and this approach warrants more study in a larger and more diverse population of subjects and activities.},
bibtype = {article},
author = {Pober, David M and Staudenmayer, John and Raphael, Christopher and Freedson, Patty S},
journal = {Medicine and science in sports and exercise},
number = {9}
}
Downloads: 0
{"_id":"FGoC7GR4uWXmaXCvq","bibbaseid":"pober-staudenmayer-raphael-freedson-developmentofnoveltechniquestoclassifyphysicalactivitymodeusingaccelerometers-2006","downloads":0,"creationDate":"2018-01-02T16:38:25.376Z","title":"Development of novel techniques to classify physical activity mode using accelerometers.","author_short":["Pober, D., M.","Staudenmayer, J.","Raphael, C.","Freedson, P., S."],"year":2006,"bibtype":"article","biburl":null,"bibdata":{"title":"Development of novel techniques to classify physical activity mode using accelerometers.","type":"article","year":"2006","identifiers":"[object Object]","keywords":"Acceleration,Activities of Daily Living,Adult,Biological,Computer-Assisted,Exercise Test,Exercise Test: classification,Exercise Test: methods,Female,Humans,Male,Markov Chains,Models,Motor Activity,Numerical Analysis,Walking","pages":"1626-34","volume":"38","websites":"http://www.ncbi.nlm.nih.gov/pubmed/16960524","month":"9","id":"3dd99bf7-92d0-39a9-a42b-0e80b43b1071","created":"2016-04-14T18:02:32.000Z","accessed":"2014-11-04","file_attached":"true","profile_id":"d28af011-5164-3af4-8522-822cff4de1eb","group_id":"e39cd875-9ef8-3fee-ad92-c9d084a63048","last_modified":"2017-10-14T23:14:07.576Z","tags":"accelerometry","read":"true","starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"Pober2006","folder_uuids":"dc34bb65-a794-457d-b88a-9687552166f4","private_publication":false,"abstract":"PURPOSE: Use of accelerometers to assess physical activity (PA) is widespread in public health research, but their utility is often limited by the accuracy of data-processing techniques. We hypothesized that more sophisticated approaches to data processing could distinguish between activity types based on accelerometer data, providing a more accurate picture of PA. METHODS: Using data from MTI Actigraphs worn by six subjects during four activities (walking, walking uphill, vacuuming, working at a computer), quadratic discriminant analysis (QDA) was performed, and a hidden Markov model (HMM) was trained to recognize the activities. The ability of the new analytic techniques to accurately classify PA was assessed. RESULTS: The mean (SE) percentage of time points for which the QDA correctly identified activity mode was 70.9% (1.2%). Computer work was correctly recognized most frequently (mean (SE) percent correct = 100% (0.01%)), followed by vacuuming (67.5% (1.5%)), uphill walking (58.2% (3.5%)), and walking (53.6% (3.3%)). The mean (SE) percentage of time points for which the HMM correctly identified activity mode was 80.8% (0.9%). Vacuuming was correctly recognized most frequently (mean (SE) percent correct = 98.8% (0.05%)), followed by computer work (97.3% (0.7%)), walking (62.6% (2.3%)), and uphill walking (62.5% (2.3%)). In contrast to a traditional method of data processing that misidentified the intensity level of 100% of the time spent vacuuming and walking uphill, the QDA and HMM approaches correctly estimated the intensity of activity 99% of the time. CONCLUSION: The novel approach of estimating activity mode, rather than activity level, may allow for more accurate field-based estimates of physical activity using accelerometer data, and this approach warrants more study in a larger and more diverse population of subjects and activities.","bibtype":"article","author":"Pober, David M and Staudenmayer, John and Raphael, Christopher and Freedson, Patty S","journal":"Medicine and science in sports and exercise","number":"9","bibtex":"@article{\n title = {Development of novel techniques to classify physical activity mode using accelerometers.},\n type = {article},\n year = {2006},\n identifiers = {[object Object]},\n keywords = {Acceleration,Activities of Daily Living,Adult,Biological,Computer-Assisted,Exercise Test,Exercise Test: classification,Exercise Test: methods,Female,Humans,Male,Markov Chains,Models,Motor Activity,Numerical Analysis,Walking},\n pages = {1626-34},\n volume = {38},\n websites = {http://www.ncbi.nlm.nih.gov/pubmed/16960524},\n month = {9},\n id = {3dd99bf7-92d0-39a9-a42b-0e80b43b1071},\n created = {2016-04-14T18:02:32.000Z},\n accessed = {2014-11-04},\n file_attached = {true},\n profile_id = {d28af011-5164-3af4-8522-822cff4de1eb},\n group_id = {e39cd875-9ef8-3fee-ad92-c9d084a63048},\n last_modified = {2017-10-14T23:14:07.576Z},\n tags = {accelerometry},\n read = {true},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Pober2006},\n folder_uuids = {dc34bb65-a794-457d-b88a-9687552166f4},\n private_publication = {false},\n abstract = {PURPOSE: Use of accelerometers to assess physical activity (PA) is widespread in public health research, but their utility is often limited by the accuracy of data-processing techniques. We hypothesized that more sophisticated approaches to data processing could distinguish between activity types based on accelerometer data, providing a more accurate picture of PA. METHODS: Using data from MTI Actigraphs worn by six subjects during four activities (walking, walking uphill, vacuuming, working at a computer), quadratic discriminant analysis (QDA) was performed, and a hidden Markov model (HMM) was trained to recognize the activities. The ability of the new analytic techniques to accurately classify PA was assessed. RESULTS: The mean (SE) percentage of time points for which the QDA correctly identified activity mode was 70.9% (1.2%). Computer work was correctly recognized most frequently (mean (SE) percent correct = 100% (0.01%)), followed by vacuuming (67.5% (1.5%)), uphill walking (58.2% (3.5%)), and walking (53.6% (3.3%)). The mean (SE) percentage of time points for which the HMM correctly identified activity mode was 80.8% (0.9%). Vacuuming was correctly recognized most frequently (mean (SE) percent correct = 98.8% (0.05%)), followed by computer work (97.3% (0.7%)), walking (62.6% (2.3%)), and uphill walking (62.5% (2.3%)). In contrast to a traditional method of data processing that misidentified the intensity level of 100% of the time spent vacuuming and walking uphill, the QDA and HMM approaches correctly estimated the intensity of activity 99% of the time. CONCLUSION: The novel approach of estimating activity mode, rather than activity level, may allow for more accurate field-based estimates of physical activity using accelerometer data, and this approach warrants more study in a larger and more diverse population of subjects and activities.},\n bibtype = {article},\n author = {Pober, David M and Staudenmayer, John and Raphael, Christopher and Freedson, Patty S},\n journal = {Medicine and science in sports and exercise},\n number = {9}\n}","author_short":["Pober, D., M.","Staudenmayer, J.","Raphael, C.","Freedson, P., S."],"urls":{"Paper":"http://bibbase.org/service/mendeley/d28af011-5164-3af4-8522-822cff4de1eb/file/76ba2f8c-cb19-44a9-29c8-1ca945c8b406/2006-Development_of_novel_techniques_to_classify_physical_activity_mode_using_accelerometers..pdf.pdf","Website":"http://www.ncbi.nlm.nih.gov/pubmed/16960524"},"bibbaseid":"pober-staudenmayer-raphael-freedson-developmentofnoveltechniquestoclassifyphysicalactivitymodeusingaccelerometers-2006","role":"author","keyword":["Acceleration","Activities of Daily Living","Adult","Biological","Computer-Assisted","Exercise Test","Exercise Test: classification","Exercise Test: methods","Female","Humans","Male","Markov Chains","Models","Motor Activity","Numerical Analysis","Walking"],"downloads":0},"search_terms":["development","novel","techniques","classify","physical","activity","mode","using","accelerometers","pober","staudenmayer","raphael","freedson"],"keywords":["acceleration","activities of daily living","adult","biological","computer-assisted","exercise test","exercise test: classification","exercise test: methods","female","humans","male","markov chains","models","motor activity","numerical analysis","walking"],"authorIDs":[]}