B Soran, L Lowes, KM Steele (2016) “Evaluation of infants with spinal muscular atrophy using convolutional neural networks.” European Conference on Computer Vision

Experimental set-up with infant positioned below Kinect depth camera.

Peer-reviewed paper at European Conference on Computer Vision:

30-second videos from a depth camera can be used in the evaluation of infants with spinal muscular atrophy.

Experimental set-up with infant positioned below Kinect depth camera.Abstract: Spinal Muscular Atrophy is the most common genetic cause of infant death. Due to its severity, there is a need for methods for automated estimation of disease progression. In this paper we propose a Convolutional-Neural-Network (CNN) model to estimate disease progression during infants’ natural behavior. With the proposed methodology, we were able to predict each child’s score on current behavior-based clinical exams with an average per-subject error of 6.96 out of 72 points (<10 % difference), using 30-second videos in leave-one-subject-out-cross-validation setting. When simple statistics were used over 30-second video-segments to estimate a score for longer videos, we obtained an average error of 5.95 (8 % error rate). By showing promising results on a small dataset (N = 70, 2-minute samples, which were handled as 1487, 30-second video segments), our methodology demonstrates that it is possible to benefit from CNNs on small datasets by proper design and data handling choices.

J Wu, BR Shuman, BW Brunton, KM Steele, JD Olson, RPN Rao (2016) “Multistep model for predicting upper-limb 3D isometric force application from pre-movement electrocorticographic features.” IEEE Engineering Medicine & Biology

Example of ECoG recording during upper-extremity force production.

Peer-reviewed paper at IEEE Engineering in Medicine & Biology Annual Conference:

Can we estimate upper-extremity force production from electrocorticographic recordings?

Example of ECoG recording during upper-extremity force production.Abstract: Neural correlates of movement planning onset and direction may be present in human electrocorticography in the signal dynamics of both motor and non-motor cortical regions. We use a three-stage model of jPCA reduced-rank hidden Markov model (jPCA-RR-HMM), regularized shrunken-centroid discriminant analysis (RDA), and LASSO regression to extract direction-sensitive planning information and movement onset in an upper-limb 3D isometric force task in a human subject. This mode achieves a relatively high true positive force-onset prediction rate of 60% within 250ms, and an above-chance 36% accuracy (17% chance) in predicting one of six planned 3D directions of isometric force using pre-movement signals. We also find direction-distinguishing information up to 400ms before force onset in the pre-movement signals, captured by electrodes placed over the limb-ipsilateral dorsal premotor regions. This approach can contribute to more accurate decoding of higher-level movement goals, at earlier timescales, and inform sensor placement. Our results also contribute to further understanding of the spatiotemporal features of human motor planning.

H Choi, TL Wren, KM Steele (2016) “Gastrocnemius operating length with ankle foot orthoses in cerebral palsy.” Prosthetics & Orthotics International

Example of gastrocnemius operating length from one subject with different AFOs.

Journal article in Prosthetics & Orthotics International:

How does the operating length of the gastrocnemius vary with different common AFOs in children with cerebral palsy?

Clinical relevance: Determining whether ankle foot orthoses stretch tight muscles can inform future orthotic design and potentially provide a platform for integrating therapy into daily life. However, stretching tight muscles must be balanced with other goals of orthoses such as improving gait and preventing bone deformities.

Choi, H. Evaluation of Gait and Muscle Function with Ankle Foot Orthoses. PhD Dissertation.

 

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, University of Washington, 2016. To download and read in full, CLICK HERE. to access through the University of Washington’s ResearchWorks Archive.

A methods figure is shown depicting an Ultrasound Coordinate System. The figure depicts how the Achilles tendon length is quantified by using an ultrasound, participant, and modeling system with motion capture markers. [In photo text: Global Coordinate System; Ultrasound Coordinate System; Figure 19 Quantification of Achilles tendon length. On left, AFO, ultrasound, and position of experimental markers. On right, scaled musculoskeletal model and schematic of coordinate systems for ultrasound to define AT length.Background: Many individuals with cerebral palsy (CP) and stroke are prescribed ankle foot orthoses
(AFOs) for use during daily life. AFOs have been shown to improve pathologic gait and walking
speed in CP and stroke by providing support and alignment. There are many different types of
AFOs available such as posterior leaf spring AFOs, rigid AFOs, and articulated AFOs. Further,
there are many parameters that can be customized or tuned for each type of AFO, such as
stiffness, heel height, shank to vertical angle, and foot plate length. However, how different types
of AFOs and the customization of specific parameters impact muscle function remains unclear.

Purpose: The goals of this dissertation were to evaluate how different types of AFOs and different
tuning parameters impact gait kinematics and muscle function. Of particular interest is the
gastrocnemius, a key muscle that crosses the knee and ankle joints and is commonly tight among
individuals with CP or stroke. Gastrocnemius operating length, defined as the total muscle and
tendon length during a functional activity, influences ankle and knee kinematics during gait.

Results/Discussion: This dissertation provides important evidence for how humans adapt to various AFO
properties and suggests important implications for the design and prescription of AFOs. This
work provides a quantitative evaluation of how AFOs impact musculotendon dynamics among
individuals with stroke (Aim 1) and cerebral palsy (Aim 2). The fabrication methods in Aim 3
creates a powerful and flexible research platform for evaluating AFO design, which may be
extended to fabrication of AFOs for daily use with further improvements in additive
manufacturing materials and methods. The final study (Aim 4), provides the first experimental
evidence combining ultrasound and musculoskeletal modeling to understand how muscle and
tendon length are impacted by AFO design. These evaluations provide guidance for future AFO
design and prescription that can not only augment human mobility for unimpaired individuals,
but also provide improve metrics for improving function and guiding rehabilitation for
individuals with neurologic impairments.