• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.305-312
• /
• 2023

Physics-Informed Neural Network (PINN) is used to invert bubble size distributions from attenuation losses. By considering a linear system for the bubble population inversion, Adaptive Learned Iterative Shrinkage Thresholding Algorithm (Ada-LISTA), which has been solved linear systems in image processing, is used as a neural network architecture in PINN. Furthermore, a regularization based on the linear system is added to a loss function of PINN and it makes a PINN have better generalization by a solution satisfying the bubble physics. To evaluate an uncertainty of bubble estimation, deep ensemble is adopted. 20 Ada-LISTAs with different initial values are trained using the same training dataset. During test with attenuation losses different from those in the training dataset, the bubble size distribution and corresponding uncertainty are indicated by average and variance of 20 estimations, respectively. Deep ensemble Ada-LISTA demonstrate superior performance in inverting bubble size distributions than the conventional convex optimization solver of CVX.

• Tuberculosis and Respiratory Diseases
• /
• v.40 no.1
• /
• pp.35-42
• /
• 1993

Background: In order to establish the etiology of the pleural effusion, routine analysis of the fluid, bacteriologic studies, cytologic tests and pleural biopsies are currently being employed. However, even with the above mentioned tests, the exact causes cannot be determined in approximately 10-20% of cases. The purpose of our study is to determine the diagnostic value of measuring ADA activity and CEA simultaneously in various pleural fluids which their etiologies have confirmed Methods: We have studied 61 cases of tuberculous pleural effusions, 17 cases of suspected tuberculous pleural effusions, 17 cases of malignant pleural effusions, 22 cases of suspected malignant pleural effusions, and 7 cases of parapneumonic pleural effusions. We have measured the ADA activity and CEA level simultaneously in pleural fluid samples in each cases. Results: 1) The ADA activity in tuberculous pleural effusion was significantly higher than that in malignant effusion. 2) The CEA level in malignant pleural effusion was significantly higher than that in tuberculous effusion. 3) With the cut-off values of the pleural fluid ADA activity more than 40 U/L and the CEA level less than 12 ng/mL, the sensitivity was 86.9%, and the specificity was 100% in the diagnosis of tuberculous effusion. With the cut-off values of the pleural fluid CEA level more than 12 g/mL and the ADA activity less than 40 U/L, the sensitivity was 76.5%, and the specificity was 100% in the diagnosis of malignant effusion. Conclusion: It is suggested that the combined assay of pleural fluid ADA activity and CEA level is very useful in the differential diagnosis of tuberculous and malignant pleural effusion.

• The Journal of the Korean dental association
• /
• v.39 no.8 s.387
• /
• pp.647-653
• /
• 2001

• Journal of KIISE:Software and Applications
• /
• v.29 no.4
• /
• pp.270-277
• /
• 2002

Text filtering is a task of deciding whether a document has relevance to a specified topic. As Internet and Web becomes wide-spread and the number of documents delivered by e-mail explosively grows the importance of text filtering increases as well. The aim of this paper is to improve the accuracy of text filtering systems by using machine learning techniques. We apply AdaBoost algorithms to the filtering task. An AdaBoost algorithm generates and combines a series of simple hypotheses. Each of the hypotheses decides the relevance of a document to a topic on the basis of whether or not the document includes a certain word. We begin with an existing AdaBoost algorithm which uses weak hypotheses with their output of 1 or -1. Then we extend the algorithm to use weak hypotheses with real-valued outputs which was proposed recently to improve error reduction rates and final filtering performance. Next, we attempt to achieve further improvement in the AdaBoost's performance by first setting weights randomly according to the continuous Poisson distribution, executing AdaBoost, repeating these steps several times, and then combining all the hypotheses learned. This has the effect of mitigating the ovefitting problem which may occur when learning from a small number of data. Experiments have been performed on the real document collections used in TREC-8, a well-established text retrieval contest. This dataset includes Financial Times articles from 1992 to 1994. The experimental results show that AdaBoost with real-valued hypotheses outperforms AdaBoost with binary-valued hypotheses, and that AdaBoost iterated with random weights further improves filtering accuracy. Comparison results of all the participants of the TREC-8 filtering task are also provided.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.3
• /
• pp.471-479
• /
• 2018

Recently, the spread of smart TV based Android iOS Set Top box has become common. This paper propose a new approach to control the TV using gestures away from the era of controlling the TV using remote control. In this paper, the AdaBoost algorithm is applied to gesture recognition by using a mono camera. First, we use Camshift-based Body tracking and estimation algorithm based on Gaussian background removal for body coordinate extraction. Using global and local feature vectors, we recognized gestures with speed change. By tracking the time interval trajectories of hand and wrist, the AdaBoost algorithm with CART algorithm is used to train and classify gestures. The principal component feature vector with high classification success rate is searched using CART algorithm. As a result, 24 optimal feature vectors were found, which showed lower error rate (3.73%) and higher accuracy rate (95.17%) than the existing algorithm.

• Food Science and Biotechnology
• /
• v.17 no.3
• /
• pp.631-636
• /
• 2008

Acanthopanax divaricatus var. albeofructus (ADA) have been shown to have various levels of activity such as antioxidant, anticancer, antivirus, and immunostimulatory effects. However, little is known about its mechanism related to the modulation of immune activities. In this study, a water extract of ADA leaves were used to treat human peripheral blood mononuclear cells (hPBMC) to determine the underlying mechanisms for the immunostimulatory effects. To characterize its immunomodulatory activity, the secretion level of various cytokines including IL-2, IL-4, IL-6, IL-10, IL-12, IFN-$\gamma$, and TNF-$\alpha$ were measured using enzyme-linked immunosorbent assay (ELISA). Treatment of hPBMC with ADA leaf extract in an in vitro experiment induced various Th1 cytokines in a dose-dependent manner. A significant increase of IL-2, IL-12, IFN-$\gamma$, and TNF-$\alpha$ secretion was observed in the presence of ADA leaf extract. In contrast, Th2 cytokines including IL-4 and IL-6 were suppressed. There was no significant change in IL-10 release. Our results showed an increase in Th1 and a decrease in Th2 cytokine secretion which suggests that ADA may influence the immune response towards a predominance of Th1 cytokines in the immune system.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.1
• /
• pp.53-60
• /
• 2014

Redundant handovers cause degraded quality of service to passengers in vehicle. This paper proposes a handover scheme suitable for users traveling in vehicles, which enables continuous learning of the handover process using a discrete-time Markov chain (DTMC). Through AdaBoost machine learning algorithm, the proposed handover scheme avoids unnecessary handover trials when a short dwell time in a target cell is expected or when the target cell is an intermediate cell through which the vehicle quickly passes. Simulation results show that the proposed scheme reduces the number of handover occurrences and maintains adequate throughput.

• The Journal of Korea Assosiation for Disability and Oral Health
• /
• v.2 no.1
• /
• pp.17-30
• /
• 2006

Federal disability law has evolved from several laws geared to protect people with disabilities since the late 1960s and early 1970s. When U.S. Congress passed the Americans with Disabilities Act (ADA) in 1990, no federal statute prohibited the majority of employers, program administrators, owners and managers of places of public accommodation and others from discriminating against people with disabilities. Toward the ends to assure equality of opportunity, full participation, independent living, and economic self-sufficiency for individuals with the disabilities, the ADA pursues three major strategies: Title I addresses inequality in employment, Title II, inequality in public services, and Title III, inequality in services and accommodations offered by private entities. The purposes of the study were to analyze the impact of the ADA on health care for persons with disabilities and to review the ongoing health policy reforms at the federal and state governments. Essential remedies that the ADA contemplates are based on two principles, simple discrimination and reasonable accommodation, which significantly improved access to quality care, especially long-term care, by persons with disabilities. However, the ongoing Medicaid policy reforms to control rising health care costs in the U.S. could threaten the access to care by persons with disabilities in optional groups and to optional care services by persons with disabilities in mandatory groups.

• Bulletin of the Korean Chemical Society
• /
• v.29 no.10
• /
• pp.1951-1959
• /
• 2008

Pentamer cluster of azodicarbonamide (ADA) based on the crystalline structure was investigated for the equilibrium structure, the stabilization energies, and the vibrational properties at various levels of the density functional theory. Stretching force constants of N${\cdot}{\cdot}{\cdot}$H or O${\cdot}{\cdot}{\cdot}$H, and angle-bending force constants of N-H${\cdot}{\cdot}{\cdot}$N or N-H${\cdot}{\cdot}{\cdot}$O for intermolecular hydrogen bonds in the pentamer cluster were obtained in 0.2-0.5 mdyn/$\AA$ and 1.6-2.0 mdyn$\AA$, respectively. The geometry of central ADA molecule fully hydrogen bonded with other four molecules shows good coincidence to the crystalline structure except the bond distances of N-H. Calculated Raman and infrared spectra of central ADA molecule in cluster represent well the experimental spectra of ADA obtained in the solid state compared to a single molecule. Detailed structural and vibrational properties of central ADA molecule in the pentamer cluster are presented.

• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.536-543
• /
• 2014

Terrain classification is still a challenging issue in image processing, especially with high resolution satellite images. The well-known obstacles include low accuracy in the detection of targets, especially for the case of man-made structures, such as buildings and roads. In this paper, we present an efficient approach to classify and detect building footprints, foliage, grass and road from high resolution grayscale satellite images. Our contribution is to build a strong classifier using AdaBoost based on a combination of co-occurrence and Haar-like features. We expect that the inclusion of Harr-like feature improves the classification performance of the man-made structures, since Haar-like feature is extracted from corner features and rectangle features. Also, the AdaBoost algorithm selects only critical features and generates an extremely efficient classifier. Experimental result indicates that the classification accuracy of AdaBoost classifier is much higher than that of the conventional classifier using back propagation algorithm. Also, the inclusion of Harr-like feature significantly improves the classification accuracy. The accuracy of the proposed method is 98.4% for the target detection and 92.8% for the classification on high resolution satellite images.