• Korean Journal of Radiology
• /
• v.2 no.4
• /
• pp.197-203
• /
• 2001

Objective: To compare the clinical utility of contrast-enhanced color Doppler US in the differentiation of retinal detachment (RD) from vitreous membrane (VM) with that of various conventional US modalities, and to analyze the enhancement patterns in cases showing an enhancement effect. Materials and Methods: In 32 eyes examined over a recent two-year period, RD (n=14) and VM (n=18) were confirmed by surgery (n=28) or clinical follow-up (n=4). In all cases, gray-scale, color Doppler, and power Doppler US were performed prior to contrast injection, and after the intravenous injection of Levovist (Schering, Berlin) by hand for 30 seconds at a dose of 2.5 g and a concentration of 300 mg/mL via an antecubital vein, contrast-enhanced color Doppler US was performed. At Doppler US, the diagnostic criterion for RD and VM was whether or not color signals were visualized in membranous structures. Results: Diagnostic accuracy was 78% at gray-scale US, 81% at color Doppler US, 59% at power Doppler US, and 97% at contrast-enhanced color Doppler US. The sensitivity of color Doppler US to color signals in RD increased from 57% to 93% after contrast enhancement. The enhancement patterns observed were signal accentuation (n=3), signal extension (n=2), signal addition (n=3), and new signal visualization (n=5). Conclusion: Contrast-enhanced color Doppler US was the most accurate US modality for differentiating RD from VM, showing a significantly increased signal detection rate in RD.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.165-170
• /
• 2014

Speech enhancement techniques that remove surrounding noise are stressed to preprocessor of speech recognition. Among the various speech enhancement techniques, Codebook-based Speech Enhancement (CBSE) operates efficiently in non-stationary noise environments. But, CBSE has some problems that inaccurate gains can be estimated if mismatch occur between input noisy signal and trained speech/noise codevectors. In this paper, the Normalized Weighting Factor (NWF) is calculated by long-term noise estimation algorithm based on Signal-to-Noise Ratio, compensated to the conventional inaccurate gains. The proposed CBSE shows better performance than conventional CBSE.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.7
• /
• pp.669-676
• /
• 2004

In this paper, a new algorithm is proposed for resolving multiple broadband incoherent sources incident on a uniform linear array. The proposed method dose not require any initial estimates for finding the transformation matrix, while the Coherent Signal-Subspace Method(CSM) proposed by Wang and Kaveh requires preliminary estimates of multigroup source location. An effective procedure is derived for finding the enhanced spectral density matrix at the center frequency using signal enhancement approach and then constructing a common signal subspace by selecting a unitary transformation matrix which is obtained via rotation of signal subspace method. The proposed approach is found to provide superior performance relative to that obtained with the CSM method in terms of sample bias of direction-of-arrival estimates.

• Speech Sciences
• /
• v.10 no.2
• /
• pp.77-84
• /
• 2003

In this research, we seek the beginning of the speech and detect the stationary speech region using lip information. Performing running average of the estimated speech signal in the stationary region, we reduce the effect of musical noise which is inherent to the conventional MlMSE (Minimum Mean Square Error) speech enhancement algorithm. In addition to it, SFM (Spectral Flatness Measure) is incorporated to reduce the speech signal estimation error due to speaking habit and some lacking lip information. The proposed algorithm with Wiener filtering shows the superior performance to the conventional methods according to MOS (Mean Opinion Score) test.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2001.06a
• /
• pp.113-116
• /
• 2001

The conventional K-means algorithm is widely used in vector quantizer design and clustering analysis. Recently modified K-means algorithm has been proposed where the codevector updating step is as fallows: new codevector = current codevector + scale factor (new centroid - current codevector). This algorithm uses a fixed value for the scale factor. In this paper, we propose a new algorithm for the enhancement of learning time in fuzzy c-means a1gorithm. Experimental results show that the proposed method produces codebooks about 5 to 6 times faster than the conventional K-means algorithm with almost the same Performance.

• Journal of Korea Multimedia Society
• /
• v.9 no.11
• /
• pp.1455-1464
• /
• 2006

In this paper, we propose an adaptive image enhancement algorithm. The proposed algorithm is classified with the MIE technique for intensity enhancement of input image and MSE techniques for saturation enhancement. The MIE technique is proposed to control the gamut mapping problem and a sudden change in image-brightness while Luminance signal is processing, The MSE techniques are proposed to control de-saturation or over-saturation while chrominance signal is processing. The proposed algorithm is focused on processing preference color for human vision in order to generate better image quality than the algorithms focused on processing uniformly to whole images, This algorithm can be applied to a monitor, TV and other display devices for high quality image.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.2
• /
• pp.104-111
• /
• 2010

This paper presents an adaptive image enhancement method using the spatial frequency property in the DCT(discrete cosine transform) compressed domain. The dc coefficients, the illumination components of image, are adjusted to compress the dynamic range of image, and the ac coefficients are modified to enhance the contrast by using the human visual system(HVS) and the spatial frequency property. The ac coefficients are separated into vertical direction, horizontal direction, and mixed spatial frequency components, and adaptively modified to minimize the block artifacts that possibly occur in the image enhancement. The proposed method using dynamic range compression and adaptive contrast enhancement shows the advanced performance without the block artifact compared with existing method.

• Journal of radiological science and technology
• /
• v.24 no.2
• /
• pp.19-22
• /
• 2001

The purpose of this study is to evaluate time course of signal enhancement on Gadomer-17 enhance MRI, and to correlate the size of enhanced area with that of the infarct area on 2'3'5'-triphenyl tetrazolium chloride(TTC) histochemical examination for the assessment of myocardial viability in reperfused Myocardial Infarction in a cat model. Tan cats(average weight: 3.8 kg) which had undergone 90 minutes of occlusion of the LAD followed by 90 minutes of reperfusion underwent MR T2-weighted imaging, and T1-weighted imaging, enhanced T1-weighted imaging. We used 1.5T Magneton Vision MRI system(Siemens, Erlangen, Germany). Signal intensities were measured in the enhanced and non-enhanced areas of enhanced T1-weighted imaging. and TTC histochemical staining the size of the abnormal signal area on each image was compared with that of the infarct area. Maximum enhancement was detected during a $40{\sim}60$ minute period with an average enhancement of $168{\pm}9.9%$ of normal myocardium. TTC staining revealed that the size of the high signal area on T2-weighted images and of the enhanced area on enhanced T1-weighted images was greater than that of the infarct area($T2=48.1%{\pm}3.7$, enhanced $T1=47.2%{\pm}2.6$, TTC $staining=38.7%{\pm}3.1$ ; p<0.05). In reperfused Myocardial Infarction in a cat model, enhanced MR imaging delineates reversibly and irreversibly damaged myocardium, with a strong enhancement and a broad temporal window. We may therefore expect that enhanced MR image is useful for demonstrating myocardial injury.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.5
• /
• pp.85-89
• /
• 2017

In this papers, we apply high performance hardware and machine learning algorithm to build an advanced VAD algorithm for speech enhancement. Since speech is made of series of phoneme, using recurrent neural network (RNN) which consider previous data is proper method to build a speech model. It is impossible to study every noise in real world. So our algorithm is builded by phoneme based study. we detect voice present frames in noisy speech signal and make enhancement of the speech signal. Phoneme based RNN model shows advanced performance in speech signal which has high correlation among each frames. To verify the performance of proposed algorithm, we compare VAD result with label data and speech enhancement result in various noise environments with previous speech enhancement algorithm.

• Korean Journal of Radiology
• /
• v.22 no.10
• /
• pp.1708-1718
• /
• 2021

Objective: The purpose of this study was to evaluate the magnetic resonance (MR) characteristics and applicability of new, uniform, extremely small iron-based nanoparticles (ESIONs) with 3-4-nm iron cores using contrast-enhanced magnetic resonance angiography (MRA). Materials and Methods: Seven types of ESIONs were used in phantom and animal experiments with 1.5T, 3T, and 4.7T scanners. The MR characteristics of the ESIONs were evaluated via phantom experiments. With the ESIONs selected by the phantom experiments, animal experiments were performed on eight rabbits. In the animal experiments, the in vivo kinetics and enhancement effect of the ESIONs were evaluated using half-diluted and non-diluted ESIONs. The between-group differences were assessed using a linear mixed model. A commercially available gadolinium-based contrast agent (GBCA) was used as a control. Results: All ESIONs showed a good T1 shortening effect and were applicable for MRA at 1.5T and 3T. The relaxivity ratio of the ESIONs increased with increasing magnetic field strength. In the animal experiments, the ESIONs showed peak signal intensity on the first-pass images and persistent vascular enhancement until 90 minutes. On the 1-week follow-up images, the ESIONs were nearly washed out from the vascular structures and organs. The peak signal intensity on the first-pass images showed no significant difference between the non-diluted ESIONs with 3-mm iron cores and GBCA (p = 1.000). On the 10-minutes post-contrast images, the non-diluted ESIONs showed a significantly higher signal intensity than did the GBCA (p < 0.001). Conclusion: In the phantom experiments, the ESIONs with 3-4-nm iron oxide cores showed a good T1 shortening effect at 1.5T and 3T. In the animal experiments, the ESIONs with 3-nm iron cores showed comparable enhancement on the first-pass images and superior enhancement effect on the delayed images compared to the commercially available GBCA at 3T.