Electrical muscle stimulation demonstrates potential for restoring functional movement and for preventing muscle atrophy after spinal cord injury (SCI). To optimize delivery of electrical stimulation protocols, an acc...
详细信息
Real time performance of non-interactive rendering of three-dimensional scenes is usually unachievable. Ray tracing is one of the methods used for rendering such scenes. The performance achieved by a sequential softwa...
详细信息
Speculative multithreading (SpMT) increases the performance by means of executing multiple threads speculatively to exploit thread-level parallelism. By combining software and hardware approaches, we have improved the...
详细信息
Speculative multithreading (SpMT) increases the performance by means of executing multiple threads speculatively to exploit thread-level parallelism. By combining software and hardware approaches, we have improved the capabilities of previous WaveScalar ISA on the basis of transactional memory system for the WaveCache Architecture. Threads are extracted at the course of static compiling, and speculatively executed as a thread-level transaction that is supported by extra hardware components, such as thread-context-table (TCT) and thread-memory-history (TMH). We have evaluated the SpMT WaveCache with 6 real benchmarks from SPEC, Mediabench and Mibench. On the whole, the SpMT WaveCache outperforms superscalar architecture ranging from 2times to 3times, and great performance gains are achieved over original WaveCache and transactional WaveCache as well.
This paper proposes the online Support Vector Regression (SVR) based value function approximation method for Reinforcement Learning (RL). This approach conserves the Support Vector Machine (SVM)'s good property, t...
详细信息
This paper proposes the online Support Vector Regression (SVR) based value function approximation method for Reinforcement Learning (RL). This approach conserves the Support Vector Machine (SVM)'s good property, the generalization which is a key issue of function approximation. Online SVR can do incremental learning and automatically track variation of environment with time-varying characteristics. Using the online SVR, we can obtain the fast and good estimation of value function and achieve RL objective efficiently. Throughout simulation tests, the feasibility and usefulness of the proposed approach is demonstrated by comparison with SARSA and Q-learning.
In this paper, we propose spatio-spectral processing techniques for the detection of dust storms and automatically finding its transport direction in 5-band NOAA-AVHRR imagery. Previous methods that use simple band ma...
详细信息
We report the design of an autonomous sensing device, which employs a thermotropic nematic liquid crystal (LC) to conduct chemical or biological sensing tasks. The development of highly reproducible methods to create ...
详细信息
We present an improved methodology for a thermal transient method enabling simultaneous measurement of thermal conductivity and specific heat of nanoscale structures with one-dimensional heat flow. The temporal respon...
详细信息
The current robot-assisted surgery has built in significant safety features, force feedback system, pre-operative planning software, but due to a lack of communication between the robot and the surrounding medical dev...
详细信息
Existing work on gait recognition is focused on casual (western) customs hence not suitable for the gulf region where long gowns are used for both men and women. This paper proposes a gait recognition solution that is...
详细信息
Existing work on gait recognition is focused on casual (western) customs hence not suitable for the gulf region where long gowns are used for both men and women. This paper proposes a gait recognition solution that is suitable for both gulf customs and casual customs. The solution is based on computing an adaptive image prediction between consecutive images. The resultant predictions are then accumulated into one image and transformed using either discrete cosine transformation (DCT) or Radon transformation. The feature vectors of the gait are computed from such transformed images. Feature modeling based on polynomial networks follows. The proposed solution is tested on a dataset with around 100 participants with mixed genders and mixed customs. The proposed system yields an impressive classification rates approaching 100% accuracy.
Reliable wireless transmission of medical data is achievable in a modern telemetry setting and is put to good use. However, intercommunications among the different medical devices and initiation of controls to avoid u...
详细信息
ISBN:
(纸本)9783642016967
Reliable wireless transmission of medical data is achievable in a modern telemetry setting and is put to good use. However, intercommunications among the different medical devices and initiation of controls to avoid unsafe incidents are generally not available. Interoperability among medical devices connected to a patient at a remote site is a challenging feature for practical realization that would increase safety. A new system is proposed to improve infusion therapy safety by implementing an interoperability feature coupled with a smart clinical decision support system. This system is designed to respond promptly to medication error alerts by the use of safety commands and interlocks. The design incorporates an innovative Intelligent Interoperability Supervisor (IIS) in the system which includes an infusion pump (or pumps) delivering drug therapy to a patient connected to relevant monitoring devices. A local decision support unit in IIS in the remote location analyzes the data transmitted by interface modules on the connected medical devices and verifies with preset patient-specific alarm criteria to detect presence of any crisis or to predict potential adverse episode. If the situation is assessed to be critical, IIS sends appropriate alert signals to a remote clinical care giver and control commands to the infusion pump to activate appropriate safety interlocks. Different critical situations of clinical cases are simulated in a conventional infusion therapy system (Configuration A), and in an IIS-implemented system (Configuration B). In Configuration A, there would be a significant delay between alarm occurrence and clinician intervention;such a time delay could lead to potential deterioration of patient condition. In Configuration B, clinically abnormal values are detected and the patient state is analyzed. If the situation demands, the system would promptly initiate the stop control command to the corresponding infusion pump ensuring safety and alerting the cl
暂无评论