Search

BACKGROUND AND PURPOSE This study was undertaken to develop a patient-centered stroke outcome measure and initial validation of the proposed Young Stroke Questionnaire (YSQ). METHODS This study assessed the reliability and discriminant validity of the YSQ. The initial questionnaire evolved from a focus group comprised of six young stroke survivors and six stroke neurologists centralized around four patient-centered domains. To determine the reliability and discriminant validity of the YSQ, 100 young stroke survivors were recruited and provided consent. Standardized clinical assessments completed included the modified Rankin Scale (mRS), National Institutes of Health Stroke Scale, and Stroke Impact Scale. Additionally, all patients were asked to complete the patient-centered YSQ. RESULTS Of the 100 enrolled patients in the study (mean age ± standard deviation = 49 ± 11.3, 58% females, 53% African American, 44% White), Cronbach alpha for all domains was >0.7. Moreover, Cronbach alpha for entire questionnaire was >0.9, indicating that the scale, with four subdomains, is internally consistent and reproducible. Discriminant validity of the scale was assessed by comparing the means of each subdomain of the YSQ among healthy subjects to the groups of stroke patients as defined by the mRS. The YSQ was able to differentiate subjects with good outcome (mRS = 0-1) from subjects with varying degree of disability as defined by the mRS (p = 0.026). CONCLUSIONS Standardized clinical assessments are not sensitive to disabilities in young stroke survivors. When compared to standardized clinical assessments, the YSQ is significantly capable of differentiating the young survivor perspective of the impact of stroke in all four subdomains.

We propose MMNet, a deep learning MIMO detection scheme that significantly outperforms existing approaches on realistic channels with the same or lower computational complexity. MMNet’s design builds on the theory of iterative soft-thresholding algorithms and uses a novel training algorithm that leverages temporal and spectral correlation in real channels to accelerate training. These innovations make it practical to train MMNet online for every realization of the channel. On spatially-correlated channels, MMNet achieves the same error rate as the next-best learning scheme (OAMPNet) at 2.5dB lower signal-to-noise ratio (SNR), and with at least 10× less computational complexity. MMNet is also 4–8dB better overall than the linear minimum mean square error (MMSE) detector.

Introduction: Stroke is a leading cause of disability affecting patients and their caregivers. Current outcome measures inadequately quantify the impact of stroke in young patients (≤65) in terms of managing education, career, and family. Methods: This study assessed the reliability and discriminate validity of the Young Stroke Questionnaire (YSQ) to further build upon the knowledge and understanding of young stroke patients and the specific needs for management and long-term outcomes relevant to young stroke survivors. The initial questionnaire evolved from a focus group comprised of 6 young stroke survivors and 6 stroke neurologists centralized around 4 patient-centered domains: work and leisure, relationships, wellbeing, and healthcare resources. To determine the reliability and discriminate validity of YSQ, 100 young stroke survivors were consented at the Neurology clinic. Standardized clinical assessments completed included the modified Rankin Scale (mRS). Additionally, all patients were asked to complete the patient-centered questionnaire, YSQ. Results: We enrolled 100 patients (Mean age ± SD = 49.2 ± 11.7, 42 Males, 53% African-American, 44% White) into the YSQ validation study. Cronbach’s α coefficients for the four domains and the total scale were calculated to measure the internal consistency of the YSQ. The α coefficients were found to be high (work & leisure α = 0.88, relationships α = 0.82, wellbeing α= 0.84, healthcare resources α= 0.75), indicating the scale with 4 subdomains is internally consistent and reproducible. The discriminant validity of the scale was assessed by comparing the means of each subdomain of YSQ among healthy subjects to the groups of stroke patients as defined by the mRS. YSQ was able to differentiate healthy subjects from subjects with varying degree of disability, as defined by the modified Rankin scale ( work and leisure p< 0.0001, relationships p< 0.0001, wellbeing p< 0.0001, Healthcare resources p= 0.017) Conclusions: Standardized clinical assessments are not sensitive to disabilities in young stroke survivors. When compared to standardized clinical assessments, the YSQ is significantly capable of differentiating the young survivor perspective of the impact of stroke in all the four subdomains.

Traditional symbol detection algorithms either perform poorly or are impractical to implement for Massive Multiple-Input Multiple-Output (MIMO) systems. Recently, several learning-based approaches have achieved promising results on simple channel models (e.g., i.i.d. Gaussian channel coefficients), but as we show, their performance degrades on real-world channels with spatial correlation. We propose MMNet, a deep learning MIMO detection scheme that significantly outperforms existing approaches on realistic channels with the same or lower computational complexity. MMNet ’s design builds on the theory of iterative soft-thresholding algorithms, and uses a novel training algorithm that leverages temporal and spectral correlation in real channels to accelerate training. These innovations make it practical to train MMNet online for every realization of the channel. On i.i.d. Gaussian channels, MMNet requires two orders of magnitude fewer operations than existing deep learning schemes but achieves near-optimal performance. On spatially-correlated channels, it achieves the same error rate as the next-best learning scheme (OAMPNet) at 2.5dB lower signal-to-noise ratio (SNR), and with at least $10\times $ less computational complexity. MMNet is also 4–8dB better overall than a classic linear scheme like the minimum mean square error (MMSE) detector.

In this paper, we present improved channel estimation and detection methods applicable for control channel in 3GPP-LTE. The physical uplink control channel (PUCCH) carries the vital information to the evolved-Node-B (eNB) such as Hybrid-ARQ (HARQ) ACK/NACK corresponding to the downlink assignment and channel quality information (CQI) for resource allocation. A robust detection of control channel helps in supporting higher data rates with low latency to the user-equipments (UEs). The PUCCH is a narrow-band code multiplexed channel and hence the inter-user interference becomes the performance limiting factor. This interference becomes severe in the presence of timing errors (leading to loss in code orthogonality) and near-far problem. The improved receiver presented in this paper employ the complex exponential functions for channel estimation and a joint minimum mean squared error (MMSE) data detection and hence are robust to timing errors and near-far problem. Also, we exploit the pilot structure for reducing the complexity of the joint channel estimation. Simulation results provided shows that the proposed receiver breaks the error floor encountered in the single user receiver.

In this paper, the performance of WiMax network with DL SIMO, DL MIMO and UL SIMO is analyzed via system simulations. Two data traffic models are considered, namely Web browsing (HTTP) and full buffer sessions, operating in multi path fading channels such as the ITU pedestrian-B and vehicular-A channels. The control channel coverage and reliability of WiMAX using realistic overhead is also analyzed in this paper.

The performance analysis of the slotted access channel for DS/SS CDMA cellular systems with and without multipath combining is presented. Three major factors that affect access channel performance are collision, blocking, and frame erasure rate (FER). All three factors can cause retransmission which results in lower throughput and longer delay. An iterative method is presented to estimate the effective arrival rate in a system without combining. The access rate sequence generated by the iteration is monotone increasing and converges if the system is stable. Using this result, a stability condition is determined in terms of the arrival rate of new messages and the FER. We modify the standard slotted ALOHA analysis so that it applies to a system with combining. The study reveals the trade-off between the messages received per slot and the FER. >