Document Detail


A universal hybrid decision tree classifier design for human activity classification.
MedLine Citation:
PMID:  23366079     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
A system that reliably classifies daily life activities can contribute to more effective and economical treatments for patients with chronic conditions or undergoing rehabilitative therapy. We propose a universal hybrid decision tree classifier for this purpose. The tree classifier can flexibly implement different decision rules at its internal nodes, and can be adapted from a population-based model when supplemented by training data for individuals. The system was tested using seven subjects each monitored by 14 triaxial accelerometers. Each subject performed fourteen different activities typical of daily life. Using leave-one-out cross validation, our decision tree produced average classification accuracies of 89.9%. In contrast, the MATLAB personalized tree classifiers using Gini's diversity index as the split criterion followed by optimally tuning the thresholds for each subject yielded 69.2%.
Authors:
Chieh Chien; Gregory J Pottie
Related Documents :
23253449 - Whole-body isometric force/torque measurements for functional assessment in neuro-rehab...
10636769 - Use of alpha-, beta- and gamma-hexachlorocyclohexane in europe, 1970-1996.
12806049 - Development of a statistical model for estimating spatial and temporal ambient ozone pa...
20962409 - Modelling of wet deposition of atmospheric polycyclic aromatic hydrocarbons by the cons...
21864359 - Validity of estimating minute-by-minute energy expenditure of continuous walking bouts ...
20382579 - An estimate of the global burden of anthropogenic ozone and fine particulate matter on ...
24165959 - Systematic bias between running speed and metabolic power data in elite soccer players:...
23004929 - Preclinical antitumor activity of thiarabine in human leukemia and lymphoma xenograft m...
22573969 - A novel energy-efficient mac aware data aggregation routing in wireless sensor networks.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference     Volume:  2012     ISSN:  1557-170X     ISO Abbreviation:  Conf Proc IEEE Eng Med Biol Soc     Publication Date:  2012  
Date Detail:
Created Date:  2013-01-31     Completed Date:  2013-08-06     Revised Date:  2014-08-21    
Medline Journal Info:
Nlm Unique ID:  101243413     Medline TA:  Conf Proc IEEE Eng Med Biol Soc     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1065-8     Citation Subset:  IM    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Accelerometry / methods
Female
Humans
Male
Models, Theoretical*
Motor Activity / physiology*
Predictive Value of Tests
Software*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Low complexity algorithm for seizure prediction using Adaboost.
Next Document:  Improving seizure detection performance reporting: analysing the duration needed for a detection.