Won Don Lee

Classifier using Extended Data Expression

C4.5 is a classification algorithm, an improved version of ID3. C4.5 is fast and deduces good result by constructing a decision tree on the problem of classification. Therefore C4.5 plays an important role in the field of classifier learning systems. This paper proposes two methods based on the decision tree for solving a classification problem. We construct the decision tree by using the measure of C4.5. First, an extended data expression of the existing C4.5 is described. Second, UChoo, a method of generating a rule from the previously made decision tree of C4.5 by using the extended data expression, is described. The rules expressed in the newly proposed method have almost the same information content as the original data set. This is quite an important result, as the size of the instance set will become usually large as the ubiquitous computation environment develops. It is not possible to keep all the individual instance data in memory. Instead, using the proposed method, large amount of data reduction can be done without losing information content

Extended simulated annealing for augmented TSP and multi-salesmen TSP

An extended simulated annealing (ESA), based on grand canonical ensemble (GCE), is proposed. An ESA is used to solve the augmented traveling salesman problems (ATSP) and the multiple traveling salesmen problems. Experimental results show that ESA has salient features such as simplicity and ability to find high-quality solutions as simulated annealing has.

An image watermarking scheme with hidden signatures

In recent years, digital watermarking has been introduced as a means of effectively protecting copyrights on digitized media such as image, audio and multimedia data. We present a watermarking technique for embedding digital seal images into a digital image, and apply this technique to each spatial and frequency domain of an image. Additionally, this technique does not require the original image to recover the seal image. This not only simplifies the encoding process but also makes up for the weak point which could be shown when an illegal attacker makes proving ownership difficult by creating a fake original. Experimental results show that the proposed watermarking technique is robust to various kinds of attacks.

Comparison of clinical characteristics between early and late patterns in hospitalized patients with ovarian hyperstimulation syndrome.

OBJECTIVE To clarify the differences in clinical characteristics between early and late ovarian hyperstimulation syndrome (OHSS). DESIGN Retrospective study. SETTING Tertiary university hospital. PATIENT(S) Ninety-four patients/cycles hospitalized for moderate-to-severe OHSS after controlled ovarian hyperstimulation (COH) for IVF/intracytoplasmic sperm injection (ICSI); early type (n = 69) and late type (n = 25). INTERVENTION(S) None. MAIN OUTCOME MEASURE(S) The COH and pregnancy outcomes, preclinical and clinical miscarriage rate, and hospital courses. RESULT(S) Serum E(2) levels (4,955.5 +/- 3,268.5 pg/mL vs. 2,340.8 +/- 960.6 pg/mL) and the number of follicles > or =11 mm on day of hCG administration (15.9 +/- 6.0 vs. 13.0 +/- 4.0), and the number of oocytes retrieved (21.9 +/- 9.7 vs. 13.2 +/- 5.9) were significantly higher in the early OHSS group compared with the late OHSS group. Clinical pregnancy rate (PR) was significantly higher in the late OHSS group (23.6% [13/55] vs. 92.0% [23/25]). There were no significant differences in multiple PR and disease severity between the two groups. CONCLUSION(S) Early OHSS is associated with excessive ovarian response to gonadotropin stimulation, whereas late OHSS is closely associated with conception cycle. Our findings do not support that late OHSS is more severe and closely associated with multiple pregnancies compared with early OHSS.

Phase Hold-up and Critical Fluidization Velocity in a Three-Phase Inverse Fluidized Bed

We studied the hydrodynamic characteristics of a three-phase inverse fluidized bed made of a transparent acrylic column of 0.115 m inner diameter and 2 m heights. Air, water and polyethylene particles were used as the gas, liquid and solid phase, respectively. We used both hydrophobic low density polyethylene (LDPE) and hydrophilic LDPE as solid phase, and distilled water as liquid phase, and filtered air as gas phase. The LDPE was chemically treated by chlorosulfonic acid to change the surface property from hydrophobic to hydrophilic. We tried to solely investigate the effect of the surface hydrophilicity of polymeric particles on the phase holdup and the critical fluidization velocity of three-phase inverse fluidization. Thus, we measured the static pressure and eventually observed critical fluidization velocity. Critical fluidization velocity became smaller in case of using MDPE hydrophobic particles than LDPE hydrophilic particles. This was thought to be due to the retardation of rising bubbles near hydrophobic particles and, subsequently, the increase of gas hold-up.

A new approach for classification of fingerprint image quality

Fingerprint identification is a technology which has been widely accepted for personal identification in many areas such as criminal investigation, access control, and Internet authentication due to its uniqueness. Most available systems for fingerprint identification use the minutiae matching for identification. The performance of minutiae extraction algorithms relies heavily on the quality of the fingerprint image because the minutiae-based approach is very sensitive to noise or image-quality degradation. Poor-quality images result in spurious and missing features which degrades the performance of the identification system. So if the quality of the image can be examined first, the image with very poor quality can be rejected. The useful information can be combined into the procedure of post-processing and matching scheme to improve the identification process. Therefore it is desirable to design a classification scheme which is able to examine the quality of a fingerprint image before it is processed by the fingerprint identification systems. Decision-tree is one of the well-known data mining methods that are used in classification problems because of its fast and effective features. In this paper a new classifier based on decision tree theory for classification of fingerprint image quality is proposed. This new classifier has many advantages in solving the fingerprint quality classification problem. It can generate the rules with almost all of the original information from the classifier even after all of the original fingerprint images are lost. It means it will be not necessary to save all the original fingerprint images because the classifier can combine the rules with the new coming data to build a more precise classifier with more information to classify the fingerprint image quality. And the new classifier can give weight to different fingerprint images if they come from the sensors with different importance. The advantages of the new classifier make the fingerprint image classification system very powerful and feasible. The proposed method has a very good performance which is proved by the experiments.

Extended Mean Field Annealing for Clustering Incomplete Data

Clustering is an very important research topic in knowledge discovery and machine learning. But sometimes the data set for clustering contains vectors missing one or more of the feature values, and is called as incomplete. The incomplete data problem exists in a wide range of field such as computer vision, biological system, and remote sensing. The problem of clustering of incomplete data is considered in this paper. In this paper, we propose a new extended MFA (mean field annealing) algorithm to solve the problem of clustering of incomplete data in the continuous-value state space and show the result of the experiment. The traditional fuzzy clustering methods calculate the centroid vectors of the clusters and then determined the membership probability, and repeat this process until the optimum solution is found. By contrast, the method proposed in this paper perturbs the membership probability, and determines whether to accept the perturbed state or not according to the changes of the energy. The result is compared with the optimal completion strategy fuzzy c-means (FCM) clustering of incomplete data algorithm and shows that the proposed method solves the problem of clustering incomplete data very well and gets a much better result.

A watermarking scheme for both spatial and frequency domain to extract the seal image without the original image

Digital watermarking has been introduced as a means of effectively protecting copyright of digitized media such as image, audio and multimedia data. We present a watermarking technique for embedding digital seal images into a digital image, and apply this technique to both the spatial and frequency domain of the image. Additionally, this technique does not require the original image to recover the seal image. This not only simplifies the encoding process but also makes up for the weak point which could be shown when an illegal attacker makes proving ownership difficult by creating a fake original. Experimental results show that the proposed watermarking technique is robust to various kinds of attacks.

In vitro maturation: Clinical applications.

Oocyte in vitro maturation (IVM) is an assisted reproductive technology in which oocytes are retrieved from the antral follicles of unstimulated or minimally stimulated ovaries. IVM of human oocytes has emerged as a promising procedure. This new technology has advantages over controlled ovarian stimulation such as reduction of costs, simplicity, and elimination of ovarian hyperstimulation syndrome. By elimination or reduction of gonadotropin stimulation, IVM offers eligible infertile couples a safe and convenient form of treatment, and IVM outcomes are currently comparable in safety and efficacy to those of conventional in vitro fertilization. IVM has been applied mainly in patients with polycystic ovary syndrome or ultrasound-only polycystic ovaries, but with time, the indications for IVM have expanded to other uncommon situations such as fertility preservation, as well as to normal responders. In this review, the current clinical experiences with IVM will be described.

Clinical application of anti-Müllerian hormone as a predictor of controlled ovarian hyperstimulation outcome

Objective In 2009 anti-Müllerian hormone (AMH) assay was approved for clinical use in Korea. This study was performed to determine the reference values of AMH for predicting ovarian response to controlled ovarian hyperstimulation (COH) using the clinical assay data. Methods One hundred sixty-two women who underwent COH cycles were included in this study. We collected data on age, basal AMH and FSH levels, total dose of gonadotropins, stimulation duration, and numbers of oocytes retrieved and fertilized. Blood samples were obtained on cycle day 3 before gonadotropin administration started. Serum AMH levels were measured at a centralized clinical laboratory center. The correlation between the AMH level and COH outcomes and cut-off values for poor and high response after COH was analyzed. Results Concentration of AMH was significantly correlated with the number of oocytes retrieved (OPU; r=0.700, p<0.001). The mean±SE serum AMH levels for poor (OPU≤3), normal (4≤OPU≤19), and high (OPU≥20) response were 0.94±0.15 ng/mL, 2.79±0.21 ng/mL, and 6.94±0.90 ng/mL, respectively. The cut-off level, sensitivity and specificity for poor and high response were 1.08 ng/mL, 85.8%, and 78.6%; and 3.57 ng/mL, 94.4%, and 83.3%, respectively. Conclusion Our data present clinical reference values of the serum AMH level for ovarian response in Korean women. The serum AMH level could be a clinically useful predictor of ovarian response to COH.