Özet:
Video Laryngeal Stroboscopy (VLS) is a stroboscopic imaging technique that allows observations of standstill and slow motion images of vibrating vocal folds. It is a valuable tool for studying vocal fold pathology, for evaluating the degrees of infiltration caused by cancerous lesions, in identifying effects of changes in pitch, tension, or intensity. It is a relatively recent technique and has gained clinical acceptance within the last two decades. The recording of vibration provides a good teaching, learning and documentation material. The resulting image data has a clear and magnified image of vocal folds allowing clinician to compare pre treatment and post treatment status of vocal folds. Besides an endoscope and a video camera, video laryngeal stroboscopic examination requires a high-intensity, pulsed light source that consists of a high-voltage, high-power supply unit, a triggering unit (TU) whose frequency can be synchronized with the fundamental frequency of the vocal folds and a Xenon flash tube. Since human vocal fold vibration fundamental frequency may reach 500Hz (for a baby), the pulsed-light source used for VLS is quite different from that used for conventional stroboscopes. In this thesis work we designed and realized a power unit and a TU for VLS. The power unit is based on a flyback topology and operates either in QR (free-running) or fixed frequency mode. To maximize efficiency, only low voltage switching is used and at light loads, frequency is reduced. The operating mode can be changed from free-running to fixed frequency mode and the regulation characteristics from primary to secondary regulation by jumpers. The output voltage can easily be adjusted by a potentiometer; thus it can also to be used in other applications like laser pumps, defibrillators, etc. The power density of the power unit is about 3W/inch3, reflecting a state-of the-art technology. It has been tested at a power level of 250W. The TU is the unit that extracts the fundamental frequency of the vocal folds for firing the xenon flash tube. Vibrations of the vocal folds are picked up by a laryngeal microphone and amplified, passed through a notch filter to reduce 50Hz pick-up, and then through a 1000Hz low pass filter and a 70Hz high pass filter in series. This filtered signal is the input of a Schmitt trigger circuit. The signal at the output of the Schmitt trigger is then inverted and integrated using two reset integrators having different time constants. The integrator outputs are compared to obtain the triggering signal for firing the Xenon flash tube. The triggering signal can easily be adjusted to obtain either standstill images of the vocal folds or slow-motion images of the vocal folds. A peak detector included within this unit can be used to determine the peak amplitude of the speech. Frequency is displayed using an ADC. The unit is optically isolated for patient safety.
