Stroke is one of the leading causes of disability worldwide. The efficacy of stroke recovery is determined by various factors, including patient adherence to their rehabilitation program. Effective communication between healthcare providers and patients is crucial for promoting patients’ adherence to rehabilitation programs. Aiming to support patient-healthcare provider communication during inpatient stroke rehabilitation, we (1) conducted semi-structured interviews with healthcare providers with expertise in inpatient stroke recovery to extract design requirements for visualizing stroke recovery progress. Using these design requirements, we (2) designed a data visualization tool representing stroke recovery. We (3) sought feedback on the visualization designs from healthcare providers and patients and integrated their feedback into the designs. Informed by the results of our studies, we provided several considerations for designing future visualization tools for patients and providers to communicate during inpatient stroke rehabilitation.

This paper introduces a novel assistive device (AD) control strategy that provides intermittent assistance and is capable of encouraging user effort during AD use. The assistance regulation controller (ARC) presented here is designed to regulate joint torque and can be used for arbitrary joint and limb motions without requiring any a priori information about the user's intended motion. Data from eight healthy subjects who performed a forearm tracking experiment using a single degree-of-freedom powered orthosis demonstrate that the ARC is capable of encouraging user effort. On average, subjects generated 40% more joint torque when performing the tracking experiment with the ARC instead of a conventional torque-amplifying controller. The results also demonstrate that the ARC is safe to use for dynamic orthosis-assisted joint motions which require several transitions into and out of the assistance and no assistance states of the controller. As such, the results also confirm that the ARC is an effective assist-as-needed control strategy suitable for use with ADs.

In this paper, we propose a new system for classifying audio segments as speech or music. The proposed system improves classification accuracy, particularly in low signal-to-noise ratio (SNR) environments. The system selects the features with the highest classification accuracy that corresponds to the SNR value. The value of this features are compared to certain thresholds, which are also adapted to the SNR. Multi-expert method of combining the features to improve classification accuracy is implemented. A new feature, termed the variance of low-band energy ratio, is also introduced. This feature produces large improvements in classification accuracy at low SNR. Performance of the proposed system is evaluated for different SNR using a library of speech and music audio segments. Using one-second segments it is shown that the proposed system can enhance the classification accuracy by 22% at SNR = -15 dB, and obtain classification accuracy of 90.3% at SNR = 0 dB.