Kristen L. Lurie

Training multi-parameter gaits to reduce the knee adduction moment with data-driven models and haptic feedback ☆

The purpose of this study was to evaluate gait retraining for reducing the knee adduction moment. Our primary objective was to determine whether subject-specific altered gaits aimed at reducing the knee adduction moment by 30% or more could be identified and adopted in a single session through haptic (touch) feedback training on multiple kinematic gait parameters. Nine healthy subjects performed gait retraining, in which data-driven models specific to each subject were determined through experimental trials and were used to train novel gaits involving a combination of kinematic changes to the tibia angle, foot progression and trunk sway angles. Wearable haptic devices were used on the back, knee and foot for real-time feedback. All subjects were able to adopt altered gaits requiring simultaneous changes to multiple kinematic parameters and reduced their knee adduction moments by 29-48%. Analysis of single parameter gait training showed that moving the knee medially by increasing tibia angle, increasing trunk sway and toeing in all reduced the first peak of the knee adduction moment with tibia angle changes having the most dramatic effect. These results suggest that individualized data-driven gait retraining may be a viable option for reducing the knee adduction moment as a treatment method for early-stage knee osteoarthritis patients with sufficient sensation, endurance and motor learning capabilities.

Haptic gait retraining for knee osteoarthritis treatment

In this paper we introduce haptic gait retraining as a new method for treating early stage medial compartment knee osteoarthritis and for reducing risk of the disease in individuals who may be susceptible. The hardware and software for implementation are presented including rotational skin stretch and vibration haptic devices used to inform subjects of alterations in gait movements. We also present a method based on real-time motion analysis for predicting new subject-specific gaits tailored to change knee joint loading. This approach uses correlation data between gait parameters and knee loading as well as a localized linearization technique to compute a final combined-parameter gait with minimum change from the subjects original, unaltered gait. Finally, we validate the haptic gait retraining system with a user experiment and show that, for the duration of the experiment, the user is able to positively change knee joint loading to approximately the same degree as HTO surgery.

Three-dimensional, distendable bladder phantom for optical coherence tomography and white light cystoscopy

Abstract. We describe a combination of fabrication techniques and a general process to construct a three-dimensional (3-D) phantom that mimics the size, macroscale structure, microscale surface topology, subsurface microstructure, optical properties, and functional characteristics of a cancerous bladder. The phantom also includes features that are recognizable in white light (i.e., the visual appearance of blood vessels), making it suitable to emulate the bladder for emerging white light+optical coherence tomography (OCT) cystoscopies and other endoscopic procedures of large, irregularly shaped organs. The fabrication process has broad applicability and can be generalized to OCT phantoms for other tissue types or phantoms for other imaging modalities. To this end, we also enumerate the nuances of applying known fabrication techniques (e.g., spin coating) to contexts (e.g., nonplanar, 3-D shapes) that are essential to establish their generalizability and limitations. We anticipate that this phantom will be immediately useful to evaluate innovative OCT systems and software being developed for longitudinal bladder surveillance and early cancer detection.

Informing haptic feedback design for gait retraining

Gait retraining, a promising treatment for knee osteoarthritis, requires the modification of three separate joint motions. In this paper we present the results of three studies to inform the design of a wearable haptic feedback system for this application. The first study motivates our choice of feedback modality for each of the motions. The latter two studies explore how to present haptic feedback to train three different motions concurrently. When feedback is presented simultaneously, subjects have poor perception of three or more haptic cues, tend to focus on only one motion at a time, and require several steps to modify all three motions. These findings suggest that vibrational feedback should be presented one joint at a time for haptic gait retraining.

Automated Mosaicing of Feature-Poor Optical Coherence Tomography Volumes With an Integrated White Light Imaging System

We demonstrate the first automated, volumetric mosaicing algorithm for optical coherence tomography (OCT) that both accommodates 6-degree-of-freedom rigid transformations and implements a bundle adjustment step amenable to generating large fields of view with endoscopic and freehand imaging systems. Our mosaicing algorithm exploits the known, rigid connection between a combined white light and OCT imaging system to reduce the computational complexity of traditional volumetric mosaicing pipelines. Specifically, the search for 3-D point correspondences is replaced by two, 2-D processing steps: We first coregister a pair of white light images in 2-D and then generate a surface map based on the volumetric OCT data, which is used to convert 2-D image homographies into 3-D volumetric transformations. A significant benefit of our dual-modality approach is its tolerance for feature-poor datasets such as bladder tissue; in contrast, approaches to mosaic feature-rich volumes with significant variations in the local intensity gradient (e.g., retinal data containing prolific vasculature) are not suitable for such feature-poor datasets. We demonstrate the performance of our algorithm using ex vivo bladder tissue and a custom tissue-mimicking phantom. The algorithm shows excellent performance over the range of volume-to-volume transformations expected during endoscopic examination and comparable accuracy with several orders of magnitude superior run times than an open-source gold-standard algorithm (N-SIFT). We anticipate the proposed algorithm can benefit bladder surveillance and surgical planning. Furthermore, its generality gives it broad applicability and potential to extend the use of OCT to clinical applications relevant to large organs typically imaged with freehand, forward-viewing endoscopes.

Variable-sized bar targets for characterizing three-dimensional resolution in OCT.

Resolution is an important figure of merit for imaging systems. We designed, fabricated and tested an optical phantom that mimics the simplicity of an Air Force Test Chart but can characterize both the axial and lateral resolution of optical coherence tomography systems. The phantom is simple to fabricate, simple to use and functions in versatile environments.

Automated, Depth-Resolved Estimation of the Attenuation Coefficient From Optical Coherence Tomography Data

We present a method for automated, depth-resolved extraction of the attenuation coefficient from Optical Coherence Tomography (OCT) data. In contrast to previous automated, depth-resolved methods, the Depth-Resolved Confocal (DRC) technique derives an invertible mapping between the measured OCT intensity data and the attenuation coefficient while considering the confocal function and sensitivity fall-off, which are critical to ensure accurate measurements of the attenuation coefficient in practical settings (e.g., clinical endoscopy). We also show that further improvement of the estimated attenuation coefficient is possible by formulating image denoising as a convex optimization problem that we term Intensity Weighted Horizontal Total Variation (iwhTV). The performance and accuracy of DRC alone and DRC+iwhTV are validated with simulated data, optical phantoms, and ex-vivo porcine tissue. Our results suggest that implementation of DRC+iwhTV represents a novel way to improve OCT contrast for better tissue characterization through quantitative imaging.

Quantitative measurements of strain and birefringence with common-path polarization-sensitive optical coherence tomography

We demonstrate the first system for optical coherence tomography (OCT) that enables simultaneous measurement of quantitative birefringence and strain in biological samples using a common-path configuration. Owing to its superior phase stability, common-path polarization sensitive optical coherence tomography (CoPPSe-OCT) achieves a sub-nanometer displacement sensitivity of 0.52 nm at an SNR of 48 dB. We utilize CoPPSe-OCT to measure reflectance, birefringence, and strain for distinguishing burnt regions in a birefringent biological sample (chicken breast muscle).

Design considerations for polarization-sensitive optical coherence tomography with a single input polarization state

Using a generalized design for a polarization-sensitive optical coherence tomography (PS-OCT) system with a single input polarization state (SIPS), we prove the existence of an infinitely large design space over which it is possible to develop simple PS-OCT systems that yield closed form expressions for birefringence. Through simulation and experiment, we validate this analysis by demonstrating new configurations for PS-OCT systems, and present guidelines for the general design of such systems in light of their inherent inaccuracies. After accounting for systemic errors, alternative designs exhibit similar performance on average to the traditional SIPS PS-OCT system. This analysis could be extended to systems with multiple input polarization states and could usher in a new generation of PS-OCT systems optimally designed to probe specific birefringent samples with high accuracy.

3D reconstruction of cystoscopy videos for comprehensive bladder records

White light endoscopy is widely used for diagnostic imaging of the interior of organs and body cavities, but the inability to correlate individual 2D images with 3D organ morphology limits its utility for quantitative or longitudinal studies of disease physiology or cancer surveillance. As a result, most endoscopy videos, which carry enormous data potential, are used only for real-time guidance and are discarded after collection. We present a computational method to reconstruct and visualize a 3D model of organs from an endoscopic video that captures the shape and surface appearance of the organ. A key aspect of our strategy is the use of advanced computer vision techniques and unmodified, clinical-grade endoscopy hardware with few constraints on the image acquisition protocol, which presents a low barrier to clinical translation. We validate the accuracy and robustness of our reconstruction and co-registration method using cystoscopy videos from tissue-mimicking bladder phantoms and show clinical utility during cystoscopy in the operating room for bladder cancer evaluation. As our method can powerfully augment the visual medical record of the appearance of internal organs, it is broadly applicable to endoscopy and represents a significant advance in cancer surveillance opportunities for big-data cancer research.