Unsupervised Cluster Analysis of Walking Activity Data for Healthy Individuals and Individuals with Lower Limb Amputation.

This is the first investigation to perform an unsupervised cluster analysis of activities performed by individuals with lower limb amputation (ILLAs) and individuals without gait impairment, in free-living conditions. Eight individuals with no gait impairments and four ILLAs wore a thigh-based accelerometer and walked on an improvised route across a variety of terrains in the vicinity of their homes. Their physical activity data were clustered to extract 'unique' groupings in a low-dimension feature space in an unsupervised learning approach, and an algorithm was created to automatically distinguish such activities. After testing three dimensionality reduction methods-namely, principal component analysis (PCA), t-distributed stochastic neighbor embedding (tSNE), and uniform manifold approximation and projection (UMAP)-we selected tSNE due to its performance and stable outputs. Cluster formation of activities via DBSCAN only occurred after the data were reduced to two dimensions via tSNE and contained only samples for a single individual. Additionally, through analysis of the t-SNE plots, appreciable clusters in walking-based activities were only apparent with ground walking and stair ambulation. Through a combination of density-based clustering and analysis of cluster distance and density, a novel algorithm inspired by the t-SNE plots, resulting in three proposed and validated hypotheses, was able to identify cluster formations that arose from ground walking and stair ambulation. Low dimensional clustering of activities has thus been found feasible when analyzing individual sets of data and can currently recognize stair and ground walking ambulation.

Human Activity Recognition of Individuals with Lower Limb Amputation in Free-Living Conditions: A Pilot Study.

This pilot study aimed to investigate the implementation of supervised classifiers and a neural network for the recognition of activities carried out by Individuals with Lower Limb Amputation (ILLAs), as well as individuals without gait impairment, in free living conditions. Eight individuals with no gait impairments and four ILLAs wore a thigh-based accelerometer and walked on an improvised route in the vicinity of their homes across a variety of terrains. Various machine learning classifiers were trained and tested for recognition of walking activities. Additional investigations were made regarding the detail of the activity label versus classifier accuracy and whether the classifiers were capable of being trained exclusively on non-impaired individuals' data and could recognize physical activities carried out by ILLAs. At a basic level of label detail, Support Vector Machines (SVM) and Long-Short Term Memory (LSTM) networks were able to acquire 77-78% mean classification accuracy, which fell with increased label detail. Classifiers trained on individuals without gait impairment could not recognize activities carried out by ILLAs. This investigation presents the groundwork for a HAR system capable of recognizing a variety of walking activities, both for individuals with no gait impairments and ILLAs.

Electromechanical response of silk fibroin hydrogel and conductive polycarbazole/silk fibroin hydrogel composites as actuator material.

Pure silk fibroin (SF) hydrogel and polycarbazole/silk fibroin (SF/PCZ) hydrogels were fabricated by solvent casting technique to evaluate electromechanical responses, dielectric properties, and cantilever deflection properties as functions of electric field strength, SF concentration, glutaraldehyde concentration, and PCZ concentration in the blends. Electromechanical properties were characterized in oscillatory shear mode at electric field strengths ranging from 0 to 600V/mm and at a temperature of 27°C. For both the pristine SF and SF/PCZ hydrogels, the storage modulus response (ΔG') and the storage modulus sensitivity (ΔG'/G'0) increased dramatically with increasing electric field strength. The pristine hydrogel possessed the highest storage modulus sensitivity value of 5.87, a relatively high value when compared with other previously studied electroactive polymers. With the addition of conductive PCZ in SF hydrogel, the storage modulus sensitivity and the relative dielectric constant decreased; the conductive polymer thus provided the softening effect under electric field. In the deflection response, the dielectrophoresis force and deflection distance increased monotonically with electric field strength, where the pure SF hydrogel showed the highest deflection distance and dielectrophoresis force.

Structure and gelation mechanism of tunable guanosine-based supramolecular hydrogels.

The mechanism of gelation of 50/50 w/w mixtures of guanosine (G) and 2',3',5'-tri-O-acetylguanosine (TAcG) in aqueous 0.354 M KCl was investigated using a combination of static light scattering (SLS), polarized and depolarized dynamic light scattering (VV and VH DLS), small-angle neutron and X-ray scattering (SANS and SAXS), and viscometric experiments. SLS and viscometry show a dramatic increase in apparent molecular weight and hydrodynamic volume at 0.2 wt % and 0.3 wt %, respectively, indicating the critical concentration for self-association of G/TAcG quartets into columnar assemblies lies below 0.2 wt %. Above this concentration, SANS and SAXS generate complementary information on the structure of the individual columnar stacks. VV and VH DLS results indicate bimodal correlation functions, whose properties suggest, respectively, translational and rotational diffusion of a bimodal distribution of particles. The fast mode appears to originate from fibrillar agglomerates of G/TAcG columnar quartet assemblies, while the slow mode comes from microgel domains. Guinier plot analysis of the SLS data probes the internal structure of the microgel domains. Collectively, the results suggest that sol and microgel phases coexist below the macroscopic gel point, and that the sol phase contains individual columnar stacks of G/TAcG quartets and fibrillar aggregates formed via lateral aggregation of these columnar assemblies. With increasing concentration, the DLS data indicate a progressive increase in the volume fraction of microgel domains, which ultimately leads to macroscopic gelation. Prior observation of a transient network contribution to the gel rheology at low temperature is attributed to the presence of individual columnar stacks within the gel network.

Tailoring the properties of guanosine-based supramolecular hydrogels.

We demonstrate that very stable hydrogels can be formed in aqueous potassium chloride solution by mixing a well-known gelator (guanosine, G) with a nongelator of similar structure (2',3',5'-tri-O-acetylguanosine, TAcG), and through a variety of characterization methods including rheology, small-angle neutron scattering, differential scanning calorimetry, and atomic force microscopy, we report substantial progress toward elucidating the factors that control the structure and stability of this fibrous gel system. The results suggest that the tailorability, long lifetime stability, and thermomechanical behavior of these gels derives from a reduction in the driving force toward crystallization with increased hydrophobic (TAcG) content, accompanied by a simultaneous decrease in fiber length and an increase in fiber width.

Hysteresis and strain hardening in the creep response of a polyaniline ER fluid.

The electrorheological creep response of PANI/silicone oil suspensions near the yield point is investigated using parallel plate rheometry. Controlled-stress, thixotropic loop experiments exhibit a pronounced hysteresis, from which we determined the static yield stress (sigma(y(static))), as the stress where onset of flow occurs on the upward part of the loop, and a dynamic yield stress (sigma(y(dynamic))), defined as the stress at which flow ceases on the downward part of the loop. The magnitude of the hysteresis, as characterized by the area under the loop, increases substantially with applied field strength and particle concentration, but decreases with increase of temperature. Consistent with literature data, the creep compliance shows an evolution from viscoelastic to viscoplastic to viscous flow behavior as the applied stress increases through the yield point. In the viscoplastic regime, the apparent equilibrium compliance, J(e)(app), shows a discrete pre-yield transition to higher values, indicating a seemingly-enhanced ductility as the applied stress nears the yield point. Measurement of the static yield stress following these creep experiments suggests that the origin of this transition is a pronounced strain-hardening effect. We conclude that strain-hardening contributes to the hysteresis observed in the thixotropic loop test.

Influence of polyethylene oxide on the rheological properties of semidilute, wormlike micellar solutions of hexadecyltrimethylammonium chloride and sodium salicylate.

The influence of polyethylene oxide (PEO) on the rheological properties of equimolar wormlike micellar solutions of hexadecyltrimethylammonium chloride (HTAC) and sodium salicylate (NaSal) is investigated, above the concentration where a micellar entanglement network is formed. PEO is known to have a temperature-dependent binding affinity for HTAC micelles. The influence of temperature, PEO concentration, and HTAC concentration is explored. Within the concentration and temperature range examined (25-100 mM HTAC and 25-50 degrees C), HTAC/NaSal solutions exhibit rheological characteristics of an entanglement network. Application of transient network theory provides information in the form of the plateau modulus, G(infinity)', the terminal viscoelastic relaxation time, tau(R), the reptation time, tau(rep), the micellar breaking time, tau(br), the mean micellar length, L , and the entanglement length, l(e). Consistent with literature data, increase of HTAC concentration results in an evolution from slow-breaking to fast-breaking behavior, accompanied by an increase in G(infinity)' and tau(rep), and decreases in tau(R), and tau(br), l(e) and L . Addition of PEO results in a substantial decrease in G(infinity)' (increase in l(e)), and corresponding increases in tau(R) and L . These observations are consistent with the idea that binding of HTAC micelles to PEO in aqueous solution decreases the number of surfactant molecules available to contribute to the entanglement network of wormlike micelles.

Polymer-surfactant complex formation and its effect on turbulent wall shear stress.

Turbulent drag reduction in Couette flow was investigated in terms of a decrease in wall shear stress for aqueous solutions of a nonionic polymer, poly(ethylene oxide) (PEO), a cationic surfactant, hexadecyltrimethylammonium chloride (HTAC), and their mixtures. Consistent with literature data, drag reduction was observed for PEO solutions above a critical molecular weight, 0.91 x 10(5) < Mc < 3.04 x 10(5) g/mol. Maximum drag reduction occurred at an optimum concentration, c(PEO)*, which scales inversely with molecular weight, and the % maximum drag reduction increases with molecular weight. For aqueous HTAC solutions, wall shear stress decreased with increasing HTAC concentration and leveled off at an optimum concentration, c(HTAC)*, comparable to the critical micelle concentration. For HTAC/PEO mixtures, the critical PEO molecular weight for drag reduction decreases, interpreted as due to an increase in hydrodynamic volume because of binding of HTAC micelles to PEO. Consistent with this interpretation, at fixed PEO concentration, maximum drag reduction was observed at an optimum HTAC concentration, c(HTAC/PEO)*, comparable to the maximum binding concentration, MBC. Also, with HTAC concentration fixed at the MBC, the optimum PEO concentration for drag reduction, c(PEO/HTAC)*, decreases relative to that, c(PEO)*, in the absence of HTAC.

Electrorheological properties of polyaniline suspensions: field-induced liquid to solid transition and residual gel structure.

Polyaniline (PANI) was synthesized via oxidative coupling polymerization in acid conditions and de-doped in solution of ammonia. The electrorheological (ER) properties of the PANI/silicone oil suspensions were investigated in oscillatory shear as functions of electric field strength, particle concentration, and host fluid viscosity. Consistent with literature, the PANI ER fluid exhibits viscoelastic behavior under the applied electric field and the ER response is strongly enhanced with increasing electric field strength and particle concentration. The dynamic moduli, G' and G'' increase dramatically, by 5 orders of magnitude, as the electric field strength is increased to 2 kV/mm. A viscoelastic liquid to solid transition occurs at a critical electric field strength, in the range Ec = 50-200 V/mm, whose value depends on particle concentration and host fluid viscosity. The fibrillar structure formed in the presence of the applied field has a static yield strength tau(y), whose value scales with electric field strength as tau(y) approximately E(1.88). When the field is switched off a residual structure remains, whose yield stress increases with the strength of the applied field and particle concentration. When the applied stress exceeds the yield stress of the residual structure, fast, fully reversible switching of the ER response is obtained.