Exposure to climate change drives stability or collapse of desert mammal and bird communities.

High exposure to warming from climate change is expected to threaten biodiversity by pushing many species toward extinction. Such exposure is often assessed for all taxa at a location from climate projections, yet species have diverse strategies for buffering against temperature extremes. We compared changes in species occupancy and site-level richness of small mammal and bird communities in protected areas of the Mojave Desert using surveys spanning a century. Small mammal communities remained remarkably stable, whereas birds declined markedly in response to warming and drying. Simulations of heat flux identified different exposure to warming for birds and mammals, which we attribute to microhabitat use. Estimates from climate projections are unlikely to accurately reflect species' exposure without accounting for the effects of microhabitat buffering on heat flux.

Online adaptive neural control of a robotic lower limb prosthesis.

OBJECTIVE: The purpose of this study was to develop and evaluate an adaptive intent recognition algorithm that continuously learns to incorporate a lower limb amputee's neural information (acquired via electromyography (EMG)) as they ambulate with a robotic leg prosthesis. APPROACH: We present a powered lower limb prosthesis that was configured to acquire the user's neural information and kinetic/kinematic information from embedded mechanical sensors, and identify and respond to the user's intent. We conducted an experiment with eight transfemoral amputees over multiple days. EMG and mechanical sensor data were collected while subjects using a powered knee/ankle prosthesis completed various ambulation activities such as walking on level ground, stairs, and ramps. Our adaptive intent recognition algorithm automatically transitioned the prosthesis into the different locomotion modes and continuously updated the user's model of neural data during ambulation. MAIN RESULTS: Our proposed algorithm accurately and consistently identified the user's intent over multiple days, despite changing neural signals. The algorithm incorporated 96.31% [0.91%] (mean, [standard error]) of neural information across multiple experimental sessions, and outperformed non-adaptive versions of our algorithm-with a 6.66% [3.16%] relative decrease in error rate. SIGNIFICANCE: This study demonstrates that our adaptive intent recognition algorithm enables incorporation of neural information over long periods of use, allowing assistive robotic devices to accurately respond to the user's intent with low error rates.

Design and preliminary testing of the RIC hybrid knee prosthesis.

We present a novel hybrid knee prosthesis that uses a motor, transmission and control system only for active dynamics tasks, while relying on a spring/damper system for passive dynamics activities. Active dynamics tasks require higher torque, lower speed, and occur less frequently than passive dynamic activities. By designing the actuation system around active tasks alone, we achieved a lightweight design (1.7 Kg w/o battery) without sacrificing peak torque (85Nm repetitive). Preliminary tests performed by an able-bodied person using a bypass orthosis show that the hybrid knee can support reciprocal stairs ambulation with low electrical energy consumption.

Analysis of using EMG and mechanical sensors to enhance intent recognition in powered lower limb prostheses.

OBJECTIVE: The purpose of this study was to determine the contribution of electromyography (EMG) data, in combination with a diverse array of mechanical sensors, to locomotion mode intent recognition in transfemoral amputees using powered prostheses. Additionally, we determined the effect of adding time history information using a dynamic Bayesian network (DBN) for both the mechanical and EMG sensors. APPROACH: EMG signals from the residual limbs of amputees have been proposed to enhance pattern recognition-based intent recognition systems for powered lower limb prostheses, but mechanical sensors on the prosthesis-such as inertial measurement units, position and velocity sensors, and load cells-may be just as useful. EMG and mechanical sensor data were collected from 8 transfemoral amputees using a powered knee/ankle prosthesis over basic locomotion modes such as walking, slopes and stairs. An offline study was conducted to determine the benefit of different sensor sets for predicting intent. MAIN RESULTS: EMG information was not as accurate alone as mechanical sensor information (p < 0.05) for any classification strategy. However, EMG in combination with the mechanical sensor data did significantly reduce intent recognition errors (p < 0.05) both for transitions between locomotion modes and steady-state locomotion. The sensor time history (DBN) classifier significantly reduced error rates compared to a linear discriminant classifier for steady-state steps, without increasing the transitional error, for both EMG and mechanical sensors. Combining EMG and mechanical sensor data with sensor time history reduced the average transitional error from 18.4% to 12.2% and the average steady-state error from 3.8% to 1.0% when classifying level-ground walking, ramps, and stairs in eight transfemoral amputee subjects. SIGNIFICANCE: These results suggest that a neural interface in combination with time history methods for locomotion mode classification can enhance intent recognition performance; this strategy should be considered for future real-time experiments.

Minimum jerk swing control allows variable cadence in powered transfemoral prostheses.

We present a novel swing phase controller for powered transfemoral prostheses based on minimum jerk theory. The proposed controller allows physiologically appropriate swing movement at any walking speed, regardless of the stance controller action. Preliminary validation in a transfemoral amputee subject demonstrates that the proposed controller provides physiological swing timing, without speed-or patient-specific tuning.

Myoelectric neural interface enables accurate control of a virtual multiple degree-of-freedom foot-ankle prosthesis.

Technological advances have enabled clinical use of powered foot-ankle prostheses. Although the fundamental purposes of such devices are to restore natural gait and reduce energy expenditure by amputees during walking, these powered prostheses enable further restoration of ankle function through possible voluntary control of the powered joints. Such control would greatly assist amputees in daily tasks such as reaching, dressing, or simple limb repositioning for comfort. A myoelectric interface between an amputee and the powered foot-ankle prostheses may provide the required control signals for accurate control of multiple degrees of freedom of the ankle joint. Using a pattern recognition classifier we compared the error rates of predicting up to 7 different ankle-joint movements using electromyographic (EMG) signals collected from below-knee, as well as below-knee combined with above-knee muscles of 12 trans-tibial amputee and 5 control subjects. Our findings suggest very accurate (5.3 ± 0.5%SE mean error) real-time control of a 1 degree of freedom (DOF) of ankle joint can be achieved by amputees using EMG from as few as 4 below-knee muscles. Reliable control (9.8 ± 0.7%SE mean error) of 3 DOFs can be achieved using EMG from 8 below-knee and above-knee muscles.

Neuromechanical sensor fusion yields highest accuracies in predicting ambulation mode transitions for trans-tibial amputees.

Advances in battery and actuator technology have enabled clinical use of powered lower limb prostheses such as the BiOM Powered Ankle. To allow ambulation over various types of terrains, such devices rely on built-in mechanical sensors or manual actuation by the amputee to transition into an operational mode that is suitable for a given terrain. It is unclear if mechanical sensors alone can accurately modulate operational modes while voluntary actuation prevents seamless, naturalistic gait. Ensuring that the prosthesis is ready to accommodate new terrain types at first step is critical for user safety. EMG signals from patient's residual leg muscles may provide additional information to accurately choose the proper mode of prosthesis operation. Using a pattern recognition classifier we compared the accuracy of predicting 8 different mode transitions based on (1) prosthesis mechanical sensor output (2) EMG recorded from residual limb and (3) fusion of EMG and mechanical sensor data. Our findings indicate that the neuromechanical sensor fusion significantly decreases errors in predicting 10 mode transitions as compared to using either mechanical sensors or EMG alone (2.3±0.7% vs. 7.8±0.9% and 20.2±2.0% respectively).

Brain-derived neurotrophic factor-estrogen interactions in the hippocampal mossy fiber pathway: implications for normal brain function and disease.

The neurotrophin brain-derived neurotrophic factor (BDNF) and the steroid hormone estrogen exhibit potent effects on hippocampal neurons during development and in adulthood. BDNF and estrogen have also been implicated in the etiology of diverse types of neurological disorders or psychiatric illnesses, or have been discussed as potentially important in treatment. Although both are typically studied independently, it has been suggested that BDNF mediates several of the effects of estrogen in the hippocampus, and that these interactions play a role in the normal brain as well as disease. Here we focus on the mossy fiber (MF) pathway of the hippocampus, a critical pathway in normal hippocampal function, and a prime example of a location where numerous studies support an interaction between BDNF and estrogen in the rodent brain. We first review the temporal and spatially regulated expression of BDNF and estrogen in the MFs, as well as their receptors. Then we consider the results of studies that suggest that 17ß-estradiol alters hippocampal function by its influence on BDNF expression in the MF pathway. We also address the hypothesis that estrogen influences the hippocampus by mechanisms related not only to the mature form of BDNF, acting at trkB receptors, but also by regulating the precursor, proBDNF, acting at p75NTR. We suggest that the interactions between BDNF and 17ß-estradiol in the MFs are potentially important in the normal function of the hippocampus, and have implications for sex differences in functions that depend on the MFs and in diseases where MF plasticity has been suggested to play an important role, Alzheimer's disease, epilepsy and addiction.

Performance of pattern recognition myoelectric control using a generic electrode grid with targeted muscle reinnervation patients.

Targeted muscle reinnervation (TMR) is a surgical technique that creates myoelectric prosthesis control sites for high-level amputees. The electromyographic signal patterns provided by the reinnervated muscles are well-suited for pattern recognition (PR) control. PR control uses more electrodes compared to conventional amplitude control techniques but their placement on the residual limb is less critical than for conventional amplitude control. In this contribution, we demonstrate that classification error and real-time control performances using a generically placed electrode grid were equivalent or superior to the performance when using targeted electrode placements on two transhumeral amputee subjects with TMR. When using a grid electrode layout, subjects were able to complete actions 0.290 sec to 1 sec faster and with greater accuracy as compared to clinically localized electrode placement (mean classification error of 1.35% and 3.2%, respectively, for a 5 movement-class classifier).These findings indicate that a grid electrode arrangement has the potential to improve control of a myoelectric prosthesis while reducing the time and effort associated with fitting the prosthesis due to clinical localization of control sites on amputee patients.

A comparison of direct and pattern recognition control for a two degree-of-freedom above elbow virtual prosthesis.

Individuals with a transhumeral amputation have a large functional deficit and require basic functions out of their prosthesis. Myoelectric prostheses have used amplitude control techniques for decades to restore one or two degrees of freedom to these patients. Pattern recognition control has also been investigated for transhumeral amputees, but in recent years, has been more focused on transradial amputees or high-level patients who have received targeted muscle reinnervation. This study seeks to use the most recent advances in pattern recognition control and investigate techniques that could be applied to the majority of the transhumeral amputee population that has not had the reinnervation surgery to determine if pattern recognition systems may provide them with improved control. In this study, able-bodied control subjects demonstrated that highly accurate two degree-of-freedom pattern recognition systems may be trained using four EMG channels. Such systems may be used to better control a prosthesis in real-time when compared to conventional amplitude control with mode switching.

Toward the development of a neural interface for lower limb prosthesis control.

Lower limb amputees form a large portion of the amputee population; however, current lower limb prostheses do not meet the needs of patients with high-level amputations who need to perform multi-joint coordinated movements. A critical missing element is an intuitive neural interface from which user intent can be determined. Surface EMG has been used as control source for upper limb prostheses for many years; for lower limb activities, however, the EMG is non-stationary and a new control strategy is required. This paper describes the work completed to date in developing a novel lower limb neural interface.

A real-time pattern recognition based myoelectric control usability study implemented in a virtual environment.

Pattern recognition based myoelectric control systems have been well researched; however very few systems have been implemented in a clinical environment. Although classification accuracy or classification error is the metric most often reported to describe how well these control systems perform, very little work research has been conducted to relate this measure to the usability of the system. This work presents a virtual clothespin usability test to assess the performance of pattern recognition based myoelectric control systems. The results suggest that users can complete the virtual task in reasonable time frames when using systems with high classification accuracies. Additionally, results indicate that a clinically-supported classifier training approach (inclusion of the transient potion of contraction signals) may reduce classification accuracy but increase real-time performance.

Principal components analysis preprocessing to reduce controller delays in pattern recognition based myoelectric control.

Information extracted from signals recorded from multi-channel surface myoelectric signal (MES) recording sites can be used as inputs to control systems for powered prostheses. For small, closely spaced muscles, such as the muscles in the forearm, the detected MES often contains contributions from more than one muscle; the contribution from each specific muscle being modified by a tissue filter between the muscle and the detection points. In this work, the measured raw MES signals are rotated by class specific rotation matrices to spatially decorrelate the measured data prior to feature extraction. This tunes the pattern recognition classifier to better discriminate the test motions. Using this preprocessing step, MES analysis windows may be cut from 256 ms to 128 ms without affecting the classification accuracy.

The effect of electrode displacements on pattern recognition based myoelectric control.

Pattern recognition based myoelectric controllers rely on a fundamental assumption that the patterns detected under a given electrode are repeatable for a given state of muscle activation. Consequently, electrode displacements on the skins surface affect the classification accuracy of the pattern based myoelectric controller. The effects of electrode displacement can be mitigated by using a training set of data which consists of patterns detected over a range of plausible displacement locations to train the control system.

Interferometric combinations of frequency-shifted mode-locked-laser pulses.

We describe some optical modulation/demodulation schemes that consist of shifting the optical frequency spectrum of mode-locked-laser pulses and interferometrically combining the shifted pulses. Several forms of frequency shift produced by electrooptic phase modulators driven at frequencies commensurable with the pulse repetition rate are considered. Analyses are made in terms of superposition of phase-modulation sidebands, and expressions are also obtained for the pulse intensity waveform in the time domain as the product of an envelope function times the original pulse waveform. Computed examples are presented.