A computer vision based candidate for functional balance test.

Balance in humans is a motor skill based on complex multimodal sensing, processing and control. Ability to maintain balance in activities of daily living (ADL) is compromised due to aging, diseases, injuries and environmental factors. Center for Disease Control and Prevention (CDC) estimate of the costs of falls among older adults was $34 billion in 2013 and is expected to reach $54.9 billion in 2020. In this paper, we present a brief review of balance impairments followed by subjective and objective tools currently used in clinical settings for human balance assessment. We propose a novel computer vision (CV) based approach as a candidate for functional balance test. The test will take less than a minute to administer and expected to be objective, repeatable and highly discriminative in quantifying ability to maintain posture and balance. We present an informal study with preliminary data from 10 healthy volunteers, and compare performance with a balance assessment system called BTrackS Balance Assessment Board. Our results show high degree of correlation with BTrackS. The proposed system promises to be a good candidate for objective functional balance tests and warrants further investigations to assess validity in clinical settings, including acute care, long term care and assisted living care facilities. Our long term goals include non-intrusive approaches to assess balance competence during ADL in independent living environments.

Source-domain spectral EEG analysis of sports-related concussion via Measure Projection Analysis.

Here, we investigated EEG-based source-level spectral differences between adolescents with sports-related concussions and healthy age matched controls. We transformed resting state EEG collected in both groups to the source domain using Independent Component Analysis (ICA) and computed the component process power spectra. For group-level analysis in the source domain, we used a probabilistic framework, Measure Projection Analysis (MPA), that has advantages over parametric k-means clustering of brain sources. MPA revealed that some frontal brain sources in the concussed group had significantly more power in the beta band (p<;0.005) and significantly less delta (p<;0.01) and theta band power (p<;0.05) than the healthy control group. These results suggest that a shift in spectral profile toward higher frequencies in some frontal brain regions might distinguish individuals with concussion from healthy controls.

Comparison of foam-based and spring-loaded dry EEG electrodes with wet electrodes in resting and moving conditions.

The introduction of dry electrodes for EEG measurements has opened up possibilities of recording EEG outside of standard clinical environments by reducing required preparation and maintenance. However, the signal quality of dry electrodes in comparison with wet electrodes has not yet been evaluated under activities of daily life (ADL) or high motion tasks. In this study, we compared the performances of foam-based and spring-loaded dry electrodes with wet electrodes under three different task conditions: resting state, walking, and cycling. Our analysis showed that signals obtained by the 2 types of dry electrodes and obtained by wet electrodes displayed high correlation for all conditions, while being prone to similar environmental and electrode-based artifacts. Overall, our results suggest that dry electrodes have a similar signal quality in comparison to wet electrodes during motion and may be more practical for use in mobile and real-time motion applications due to their convenience. In addition, we conclude that as with wet electrodes, post-processing can mitigate motion artifacts in ambulatory EEG acquisition.

Basis selection for maximally independent EEG sources.

We suggest a solution to the following problem: "Given multichannel linear source mixture data Y, and an overcomplete dictionary, A, of source projections, ai, how can we construct a complete basis, A0, by selecting columns from A such that the sources X = A0(-1)Y are as statistically independent as possible from each other?". While conventional independent component analysis (ICA) methods learn the mixing matrix A0 from scratch given Y, we restrict ourselves to selecting basis vectors from a known overcomplete dictionary. We develop two methods based on modifications of the maximum likelihood equivalent of the Infomax approach and the reconstruction-ICA (RICA) algorithm. We show that on realistic synthetic electroencephalographic (EEG) data our algorithms can find the true sources in the case of a highly coherent dictionary while requiring relatively fewer data points compared to other algorithms. On real EEG data, our algorithms obtain higher mutual information reduction.