Projects

In our review of three publicly available national datasets—

• • the American Community Survey (ACS),

  • the National Household Transportation Survey (NHTS)

  • the American Time Use Survey (ATUS)

we identified several factors that consistently differentiate travel behavior between people with disabilities and people without disabilities utilizing unsupervised clustering approaches. 

• These datasets were developed from nationwide survey responses that contain data on respondents’ travel patterns, demographic information, and technology use.

The research aims of this study involve:

• Statistical phenotyping (clustering) of chronic low back pain (cLBP) patients using baseline demographics, physical exam assessments, patient-reported outcomes, biomechanical features, and biological data

• Examination of cLBP phenotypes for key characteristics associated with the clinical outcomes of pain, disability, and function

• Identification of patient characteristics associated with different treatment patterns. We are currently using a variety of supervised and unsupervised machine learning approaches, such as k-means and hierarchical clustering, group-based trajectory modeling, support-vector machines, and neural networks.

This aim of this study was to demonstrate the benefits of seat elevating devices on power wheelchairs while controlling for other variables. Data was collected from the Functional Mobility Assessment (FMA). A  measure that assesses patients satisfaction and ability to carry out specific tests such a transferring and reaching. 

The purpose of this study was to examine effects of perturbations to specific cortical areas on the control of unconstrained reaches against gravity by younger and older adults. Specific statistical techniques include generalized linear modeling (GLM).

The purpose of this project was to determine the reliability and validity of the modified Heckmatt Scale in assessing muscle echotexture in spasticity.