A hierarchical ensemble model for automated assessment of stroke impairment

Assessment of sensory, motor and cognitive function of stroke subjects provide important information to guide patient rehabilitation. As many of the currently used measures are inherently subjective and use course rating scales, here we propose a hierarchical ensemble network that can automatically identify stroke patients and assess their upper limb functionality objectively, based on experimental task data. We compare our neural network ensemble model with ten combinations of different classifiers and ensemble schemes, showing that it significantly outperforms competitors. We also demonstrate that our measure scale is congruent with clinical information, responsive with changes of patients motor function, and reliable in terms of test-retest configuration.

[1]  Paul Cisek,et al.  Kinematics and kinetics of multijoint reaching in nonhuman primates. , 2003, Journal of neurophysiology.

[2]  Pedro M. Domingos,et al.  Beyond Independence: Conditions for the Optimality of the Simple Bayesian Classifier , 1996, ICML.

[3]  A P Georgopoulos,et al.  On the relations between the direction of two-dimensional arm movements and cell discharge in primate motor cortex , 1982, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[4]  Anthony Widjaja,et al.  Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond , 2003, IEEE Transactions on Neural Networks.

[5]  Stephen H Scott,et al.  Limited transfer of learning between unimanual and bimanual skills within the same limb , 2006, Nature Neuroscience.

[6]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[7]  Stephen H. Scott,et al.  Apparatus for measuring and perturbing shoulder and elbow joint positions and torques during reaching , 1999, Journal of Neuroscience Methods.

[8]  S. Scott,et al.  A motor learning strategy reflects neural circuitry for limb control , 2003, Nature Neuroscience.

[9]  Susanne Iwarsson,et al.  Indicators for return to work after stroke and the importance of work for subjective well-being and life satisfaction. , 2003, Journal of rehabilitation medicine.

[10]  S. Black,et al.  The Fugl-Meyer Assessment of Motor Recovery after Stroke: A Critical Review of Its Measurement Properties , 2002, Neurorehabilitation and neural repair.

[11]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[12]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[13]  John Hallam,et al.  IEEE International Joint Conference on Neural Networks , 2005 .

[14]  K. Furie,et al.  Heart disease and stroke statistics--2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. , 2007, Circulation.