http://hdl.handle.net/10397/1707 Search the Channel search http://repository.lib.polyu.edu.hk/jspui/simple-search http://hdl.handle.net/10397/5617 Title: Cardiac parameters analysis for zebrafish heart regeneration based on highfrequency ultrasound imaging<br/><br/>Authors: Yu, Yanyan; Lien, Ching-Ling; Shung, K. K.; Sun, Lei<br/><br/>Abstract: Zebrafish can fully regenerate their myocardium after up to 20% ventricular resection within 2 months without evidence of scar tissues. The extraordinary regenerative abilities provide a significant model system to study the activation of the regenerative potential of human heart tissue. In order to characterize cardiac functions of zebrafish during heart regeneration, we used high-frame-rate high-frequency ultrasound system with the capabilities of 75 MHz B-mode imaging to monitor real-time cardiac parameters. Longitudinal in vivo experiments was carried out to capture echocardiographic images from individual fishes. A pilot study on 6 fish over 30 days post amputation (dpa) was performed using this technique. A total of over 400 videos were captured. To process the large video data, an automatic image segmentation algorithm was developed. By using information obtained from temporal decorrelation of the B-mode image sequence, the epicardium is determined in a region-based level set framework combined with shape priors and Kalman filtering. Afterwards, the size of the heart was estimated frame by frame using the outlined epicardium to obtain its dynamic variation. Subsequently, the maximum and minimum of the heart size was used to calculate ejection fraction (EF). The time course of the mean EF (n=6) with a "V" shape indicates the strong ability of zebrafish to recovery cardiac functions along its morphological regeneration.<br/><br/>Description: DOI: 10.1109/ULTSYM.2010.5935684 http://hdl.handle.net/10397/5616 Title: Particle trapping for acoustic tweezers<br/><br/>Authors: Yu, Yanyan; Qiu, Weibao; Sun, Lei<br/><br/>Abstract: The optical tweezers has been found to have many biomedical applications in trapping macromolecules and cells. A recent theoretical study has shown that under appropriate conditions acoustic trapping is also possible [1]. Compared to the optical tweezers, the acoustic tweezers is more useful in light opaque media. In this paper, first we present the range where a particle can be trapped in single focused field along the axial direction from a 100MHz concave circular transducer on various sizes of particles. Secondly, we also proposed a multitrap model of acoustic tweezers which can trap 4 particles simultaneously. A 100MHz 2-D phased array with each line composed of 80 elements was used to generate and control the multiple-focus acoustic field. Both of the two fields generated by concave circular and phased array transducer were evaluated based on finite-element model. The radiation force was computed by the momentum transfer occurs between the mediums inside and outside the particle according to the law of conservation. The result demonstrates that the acoustic tweezers not only can manipulate larger particles, but also owns the feasibility of multitrap.<br/><br/>Description: DOI: 10.1109/ULTSYM.2011.0380 http://hdl.handle.net/10397/5615 Title: A programmable, cost-effective, real-time high frequency ultrasound imaging board based on high-speed FPGA<br/><br/>Authors: Qiu, Weibao; Yu, Yanyan; Sun, Lei<br/><br/>Abstract: High frequency (>20MHz) ultrasound imaging has made it possible to delineate small structures with fine spatial resolution. Previously, a high frequency high frame rate imaging system had been developed for small animal cardiac imaging. This paper presents further development of a programmable, low-cost, real-time high frequency ultrasound imaging board based on high-speed FPGA. Utilization of FPGA facilitates programmable applications with high flexibility. The printed circuit board (PCB) design achieves cost-effectiveness and compactness. The PCI bus interface allows high speed data transfer and real-time imaging. The dedicated front-end electronics showed a minimum detectable signal of 18μV, allowing a 50dB dynamic range at a total gain of 50dB. The high-speed analog to digital converter (ADC) with a typical 10.8 bit ENOB was employed to accomplish high speed data acquisition. Algorithms such as band-pass filter, envelope detection and digital scan converter were implemented in the FPGA with high speed and high flexibility. Finally, wire phantom experiment showed good performance of such a programmable and compact design.<br/><br/>Description: DOI: 10.1109/ULTSYM.2010.5935574 http://hdl.handle.net/10397/1828 Title: Robot-assisted wrist training for chronic stroke : a comparison between electromyography (EMG) driven robot and passive motion<br/><br/>Authors: Hu, Xiaoling; Tong, Kai-yu Raymond; Song, Rong; Zheng, X. J.; Lui, Ka-him; Ng, Sheung-mei Shamay; Au-Yeung, Suk-yin Stephanie; Leung, Woon-fong Wallace<br/><br/>Abstract: The motor recovery procedure during robot-assisted wrist rehabilitation for persons after stroke has not been well studied previously. In this work, we carried out acomparative study on the training effects on 10 hemiplegic persons with chronic stroke between a wrist treatment assisted by an electromyography (EMG)-driven robotic system(interactive treatment, n=5, EMG group) and a wrist treatment assisted by a clinical robot system with continuous passive motion (n=5, passive group). Significant decreases (P<0.05) in muscle spasticity were observed at the wrist joint in both the EMG and passive groups; and reduced muscle spasticity at the elbow joint were also obtained in the EMG group (P<0.05). These spasticity decreases were associated with the reduction of EMG activation levels during the training. The EMG-driven robot-assisted training also improved the muscle coordination capability of the persons after stroke.