-
[show abstract]
[hide abstract]
ABSTRACT: Computer speech recognition of individuals with dysarthria, such as cerebral palsy patients requires a robust technique that can handle conditions of very high variability and limited training data. In this study, application of a 10 state ergodic hidden Markov model (HMM)/artificial neural network (ANN) hybrid structure for a dysarthric speech (isolated word) recognition system, intended to act as an assistive tool, was investigated. A small size vocabulary spoken by three cerebral palsy subjects was chosen. The effect of such a structure on the recognition rate of the system was investigated by comparing it with an ergodic hidden Markov model as a control tool. This was done in order to determine if this modified technique contributed to enhanced recognition of dysarthric speech. The speech was sampled at 11 kHz. Mel frequency cepstral coefficients were extracted from them using 15 ms frames and served as training input to the hybrid model setup. The subsequent results demonstrated that the hybrid model structure was quite robust in its ability to handle the large variability and non-conformity of dysarthric speech. The level of variability in input dysarthric speech patterns sometimes limits the reliability of the system. However, its application as a rehabilitation/control tool to assist dysarthric motor impaired individuals holds sufficient promise.
Medical Engineering & Physics 11/2006; 28(8):741-8. · 1.62 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: In this study, a hidden Markov Model was constructed and conditions were investigated that would provide improved performance for a dysarthric speech (isolated word) recognition system. The speaker dependant system was intended to act as an assistive/control tool. A small size vocabulary spoken by three cerebral palsy subjects was chosen. Fast Fourier transform, linear predictive, and Mel frequency cepstral coefficients extracted from data provided training input to several whole-word hidden Markov model configurations. The effect of model structure, number of states, and frame rates were also investigated. It was noted that a 10-state ergodic model using 15 msec frames was better than other configurations. Furthermore, it was found that a Mel cepstrum based model outperformed a fast Fourier transform and linear prediction based model. The system offers effective and robust application as a rehabilitation and/or control tool to assist dysarthric motor impaired individuals.
IEEE Transactions on Neural Systems and Rehabilitation Engineering 01/2006; 13(4):558-61. · 3.44 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Rotary ventricular assist devices (VADs) are primarily used during surgery and short-term postoperative assistance. There are issues involving the use of rotary VADs including the behavior of flow emanating from the assist device which might induce flow irregularities that could potentially lead to stroke or cerebrovascular accident. An in vitro flow visualization study using particle image velocimetry was conducted to evaluate flow fields at the bend of outlet cannulae with angles ranging from 30 degrees to 75 degrees. A multiple disk centrifugal blood pump produced steady, laminar flow fields at the outlet. Flow fields at the bend of each cannula demonstrated the development of a boundary layer along the inner wall and shift of fluid momentum to the outer wall. Stagnant zones along the inner wall were prevalent for each cannula and increased in severity with increasing angle. The outlet flow fields of rotary VADs have the potential to produce thrombus, feasibly leading to thromboembolisms should the bend angle of the outlet cannula be acute enough. Therefore, the placement of the outlet cannula or conduit should be taken into consideration during implantation.
Artificial Organs 09/2002; 26(8):714-23. · 2.00 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: A multiple disk centrifugal pump (MDCP) is under investigation as a potential left ventricular assist device. As is the case with most shaft driven pumps, leakage problems around the shaft/shaft seal interface are of major interest. If leakage were to occur during or after implantation, potential events such as blood loss, clotting, blood damage, and/or infections might result in adverse effects for the patient. Because these effects could be quite disastrous, potential shaft and shaft seal materials have been investigated to determine the most appropriate course to limit these effects. Teflon and nylon shaft seals were analyzed as potential candidates along with a stainless steel shaft and a Melonite coated shaft. The materials and shafts were evaluated under various time durations (15, 30, 45, and 60 min), motor speeds (800, 1,000, 1,200, and 1, 400 rpm), and outer diameters (½ and ¾ inches). The motor speed and geometrical configurations were typical for the MDCP under normal physiologic conditions. An air and water study was conducted to analyze the inner diameter wear, the inner temperature values, and the outer temperature values. Statistical comparisons were computed for the shaft seal materials, the shafts, and the outer diameters along with the inner and outer temperatures. The conclusions made from the results indicate that both the tested shaft seal materials and shaft materials are not ideal candidates to be used for the MDCP. Teflon experienced a significant amount of wear in air and water studies. Nylon did experience little wear, but heat generation was an evident problem. A water study on nylon was not conducted because of its molecular structure.
Artificial Organs 05/1999; 23(6):552 - 558. · 2.00 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Computer speech recognition of individuals with dysarthria, such as cerebral palsy patients, requires a robust technique that can handle conditions of very high variability and limited training data. In this study, a hidden Markov model (HMM) was constructed and conditions investigated that would provide improved performance for a dysarthric speech (isolated word) recognition system intended to act as an assistive/control tool. In particular, we investigated the effect of high-frequency spectral components on the recognition rate of the system to determine if they contributed useful additional information to the system. A small-size vocabulary spoken by three cerebral palsy subjects was chosen. Mel-frequency cepstral coefficients extracted with the use of 15 ms frames served as training input to an ergodic HMM setup. Subsequent results demonstrated that no significant useful information was available to the system for enhancing its ability to discriminate dysarthric speech above 5.5 kHz in the current set of dysarthric data. The level of variability in input dysarthric speech patterns limits the reliability of the system. However, its application as a rehabilitation/control tool to assist dysarthric motor-impaired individuals such as cerebral palsy subjects holds sufficient promise.
The Journal of Rehabilitation Research and Development 42(3):363-71. · 1.78 Impact Factor
